Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems. Sébastien Bubeck Theory Group
|
|
- Lucas Page
- 5 years ago
- Views:
Transcription
1 Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems Sébastien Bubeck Theory Group
2 Part 1: i.i.d., adversarial, and Bayesian bandit models
3 i.i.d. multi-armed bandit, Robbins [1952]
4 i.i.d. multi-armed bandit, Robbins [1952] Known parameters: number of arms n and (possibly) number of rounds T n.
5 i.i.d. multi-armed bandit, Robbins [1952] Known parameters: number of arms n and (possibly) number of rounds T n. Unknown parameters: n probability distributions ν 1,..., ν n on [0, 1] with mean µ 1,..., µ n (notation: µ = max i [n] µ i ).
6 i.i.d. multi-armed bandit, Robbins [1952] Known parameters: number of arms n and (possibly) number of rounds T n. Unknown parameters: n probability distributions ν 1,..., ν n on [0, 1] with mean µ 1,..., µ n (notation: µ = max i [n] µ i ). Protocol: For each round t = 1, 2,..., T, the player chooses I t [n] based on past observations and receives a reward/observation Y t ν It (independently from the past).
7 i.i.d. multi-armed bandit, Robbins [1952] Known parameters: number of arms n and (possibly) number of rounds T n. Unknown parameters: n probability distributions ν 1,..., ν n on [0, 1] with mean µ 1,..., µ n (notation: µ = max i [n] µ i ). Protocol: For each round t = 1, 2,..., T, the player chooses I t [n] based on past observations and receives a reward/observation Y t ν It (independently from the past). Performance measure: The cumulative regret is the difference between the player s accumulated reward and the maximum the player could have obtained had she known all the parameters, R T = T µ E Y t. t [T ] Fundamental tension between exploration and exploitation. Many applications!
8 i.i.d. multi-armed bandit: fundamental limitations How small can we expect R T to be? Consider the 2-armed case where ν 1 = Ber(1/2) and ν 2 = Ber(1/2 + ξ ) where ξ { 1, 1} is unknown.
9 i.i.d. multi-armed bandit: fundamental limitations How small can we expect R T to be? Consider the 2-armed case where ν 1 = Ber(1/2) and ν 2 = Ber(1/2 + ξ ) where ξ { 1, 1} is unknown. With τ expected observations from the second arm there is a probability at least exp( τ 2 ) to make the wrong guess on the value of ξ.
10 i.i.d. multi-armed bandit: fundamental limitations How small can we expect R T to be? Consider the 2-armed case where ν 1 = Ber(1/2) and ν 2 = Ber(1/2 + ξ ) where ξ { 1, 1} is unknown. With τ expected observations from the second arm there is a probability at least exp( τ 2 ) to make the wrong guess on the value of ξ. Let τ(t) be the expected number of pulls of arm 2 up to time t when ξ = 1.
11 i.i.d. multi-armed bandit: fundamental limitations How small can we expect R T to be? Consider the 2-armed case where ν 1 = Ber(1/2) and ν 2 = Ber(1/2 + ξ ) where ξ { 1, 1} is unknown. With τ expected observations from the second arm there is a probability at least exp( τ 2 ) to make the wrong guess on the value of ξ. Let τ(t) be the expected number of pulls of arm 2 up to time t when ξ = 1. R T (ξ = +1) + R T (ξ = 1) τ(t ) + T exp( τ(t) 2 ) t=1 min t [T ] (t + T exp( t 2 )) log(t 2 ). See Bubeck, Perchet and Rigollet [2012] for the details.
12 i.i.d. multi-armed bandit: fundamental limitations How small can we expect R T to be? Consider the 2-armed case where ν 1 = Ber(1/2) and ν 2 = Ber(1/2 + ξ ) where ξ { 1, 1} is unknown. With τ expected observations from the second arm there is a probability at least exp( τ 2 ) to make the wrong guess on the value of ξ. Let τ(t) be the expected number of pulls of arm 2 up to time t when ξ = 1. R T (ξ = +1) + R T (ξ = 1) τ(t ) + T exp( τ(t) 2 ) t=1 min t [T ] (t + T exp( t 2 )) log(t 2 ). See Bubeck, Perchet and Rigollet [2012] for the details. For fixed the lower bound is log(t ), and for the worse ( 1/ T ) it is T (Auer, Cesa-Bianchi, Freund and Schapire [1995]: Tn for the n-armed case).
13 i.i.d. multi-armed bandit: fundamental limitations Notation: i = µ µ i and N i (t) is the number of pulls of arm i up to time t. Then one has R T = n i=1 ien i (T ).
14 i.i.d. multi-armed bandit: fundamental limitations Notation: i = µ µ i and N i (t) is the number of pulls of arm i up to time t. Then one has R T = n i=1 ien i (T ). For p, q [0, 1], kl(p, q) := p log p q + (1 p) log 1 p 1 q.
15 i.i.d. multi-armed bandit: fundamental limitations Notation: i = µ µ i and N i (t) is the number of pulls of arm i up to time t. Then one has R T = n i=1 ien i (T ). For p, q [0, 1], kl(p, q) := p log p q + (1 p) log 1 p 1 q. Theorem (Lai and Robbins [1985]) Consider a strategy s.t. a > 0, we have EN i (T ) = o(t a ) if i > 0. Then for any Bernoulli distributions, R T lim inf T + log(t ) i: i >0 i kl(µ i, µ ).
16 i.i.d. multi-armed bandit: fundamental limitations Notation: i = µ µ i and N i (t) is the number of pulls of arm i up to time t. Then one has R T = n i=1 ien i (T ). For p, q [0, 1], kl(p, q) := p log p q + (1 p) log 1 p 1 q. Theorem (Lai and Robbins [1985]) Consider a strategy s.t. a > 0, we have EN i (T ) = o(t a ) if i > 0. Then for any Bernoulli distributions, R T lim inf T + log(t ) i: i >0 1 Note that 2 i i kl(µ i,µ ) µ (1 µ ) 2 i the Lai and Robbins lower bound is i: i >0 i kl(µ i, µ ). so up to a variance-like term log(t ) 2 i.
17 i.i.d. multi-armed bandit: fundamental strategy Hoeffding s inequality: w.p. 1 1/T, t [T ], i [n], µ i 1 2 log(t ) Y s + =: UCB i (t). N i (t) N i (t) s<t:i s=i
18 i.i.d. multi-armed bandit: fundamental strategy Hoeffding s inequality: w.p. 1 1/T, t [T ], i [n], µ i 1 2 log(t ) Y s + =: UCB i (t). N i (t) N i (t) s<t:i s=i UCB (Upper Confidence Bound) strategy (Lai and Robbins [1985], Agarwal [1995], Auer, Cesa-Bianchi and Fischer [2002]): I t argmax UCB i (t). i [n]
19 i.i.d. multi-armed bandit: fundamental strategy Hoeffding s inequality: w.p. 1 1/T, t [T ], i [n], µ i 1 2 log(t ) Y s + =: UCB i (t). N i (t) N i (t) s<t:i s=i UCB (Upper Confidence Bound) strategy (Lai and Robbins [1985], Agarwal [1995], Auer, Cesa-Bianchi and Fischer [2002]): I t argmax UCB i (t). i [n] Simple analysis: on a 1 2/T probability event one has N i (t) 8 log(t )/ 2 i UCB i (t) < µ UCB i (t),
20 i.i.d. multi-armed bandit: fundamental strategy Hoeffding s inequality: w.p. 1 1/T, t [T ], i [n], µ i 1 2 log(t ) Y s + =: UCB i (t). N i (t) N i (t) s<t:i s=i UCB (Upper Confidence Bound) strategy (Lai and Robbins [1985], Agarwal [1995], Auer, Cesa-Bianchi and Fischer [2002]): I t argmax UCB i (t). i [n] Simple analysis: on a 1 2/T probability event one has N i (t) 8 log(t )/ 2 i UCB i (t) < µ UCB i (t), so that EN i (T ) log(t )/ 2 i and in fact R T log(t ). i i: i >0
21 i.i.d. multi-armed bandit: going further 1. Optimal constant (replacing 8 by 1/2 in the UCB regret bound) and Lai and Robbins variance-like term (replacing i by kl(µ i, µ )): see Cappé, Garivier, Maillard, Munos and Stoltz [2013].
22 i.i.d. multi-armed bandit: going further 1. Optimal constant (replacing 8 by 1/2 in the UCB regret bound) and Lai and Robbins variance-like term (replacing i by kl(µ i, µ )): see Cappé, Garivier, Maillard, Munos and Stoltz [2013]. 2. In many applications one is merely interested in finding the best arm (instead of maximizing cumulative reward): this is the best arm identification problem. For the fundamental strategies see Even-Dar, Mannor and Mansour [2006] for the fixed-confidence setting (see also Jamieson and Nowak [2014] for a recent short survey) and Audibert, Bubeck and Munos [2010] for the fixed budget setting. Key takeaway: one needs of order H := i 2 i rounds to find the best arm.
23 i.i.d. multi-armed bandit: going further 1. Optimal constant (replacing 8 by 1/2 in the UCB regret bound) and Lai and Robbins variance-like term (replacing i by kl(µ i, µ )): see Cappé, Garivier, Maillard, Munos and Stoltz [2013]. 2. In many applications one is merely interested in finding the best arm (instead of maximizing cumulative reward): this is the best arm identification problem. For the fundamental strategies see Even-Dar, Mannor and Mansour [2006] for the fixed-confidence setting (see also Jamieson and Nowak [2014] for a recent short survey) and Audibert, Bubeck and Munos [2010] for the fixed budget setting. Key takeaway: one needs of order H := i 2 i rounds to find the best arm. 3. The UCB analysis extends to sub-gaussian reward distributions. For heavy-tailed distributions, say with 1 + ε moment for some ε (0, 1], one can get a regret that scales with 1/ε (instead of 1 ) by using a robust mean i i estimator, see Bubeck, Cesa-Bianchi and Lugosi [2012].
24 Adversarial multi-armed bandit, Auer, Cesa-Bianchi, Freund and Schapire [1995, 2001] For t = 1,..., T, the player chooses I t [n] based on previous observations, and simultaneously an adversary chooses a loss vector l t [0, 1] n. The player s loss/observation is l t (I t ).
25 Adversarial multi-armed bandit, Auer, Cesa-Bianchi, Freund and Schapire [1995, 2001] For t = 1,..., T, the player chooses I t [n] based on previous observations, and simultaneously an adversary chooses a loss vector l t [0, 1] n. The player s loss/observation is l t (I t ). The regret and pseudo-regret are defined as: R T = max (l t (I t ) l t (i)), R T = max E (l t (I t ) l t (i)). i [n] i [n] t [T ] t [T ]
26 Adversarial multi-armed bandit, Auer, Cesa-Bianchi, Freund and Schapire [1995, 2001] For t = 1,..., T, the player chooses I t [n] based on previous observations, and simultaneously an adversary chooses a loss vector l t [0, 1] n. The player s loss/observation is l t (I t ). The regret and pseudo-regret are defined as: R T = max (l t (I t ) l t (i)), R T = max E (l t (I t ) l t (i)). i [n] i [n] t [T ] t [T ] Obviously ER T R T and there is equality in the oblivious case ( adversary s choice are independent of the player s choice). The case where l 1,..., l T is an i.i.d. sequence corresponds to the i.i.d. case we just studied. In particular we have a Tn lower bound.
27 Adversarial multi-armed bandit, fundamental strategy Exponential weights strategy for full information (l t is observed at the end of round t): play I t at random from p t where p t+1 (i) = 1 Z t+1 p t (i) exp( ηl t (i)).
28 Adversarial multi-armed bandit, fundamental strategy Exponential weights strategy for full information (l t is observed at the end of round t): play I t at random from p t where p t+1 (i) = 1 Z t+1 p t (i) exp( ηl t (i)). In five lines one can show R T 2T log(n) with p 1 (i) = 1/n:
29 Adversarial multi-armed bandit, fundamental strategy Exponential weights strategy for full information (l t is observed at the end of round t): play I t at random from p t where p t+1 (i) = 1 Z t+1 p t (i) exp( ηl t (i)). In five lines one can show R T 2T log(n) with p 1 (i) = 1/n: Ent(δ j p t ) Ent(δ j p t+1 ) = log p t+1(j) p t (j) = log 1 Z t+1 ηl t (j)
30 Adversarial multi-armed bandit, fundamental strategy Exponential weights strategy for full information (l t is observed at the end of round t): play I t at random from p t where p t+1 (i) = 1 Z t+1 p t (i) exp( ηl t (i)). In five lines one can show R T 2T log(n) with p 1 (i) = 1/n: Ent(δ j p t ) Ent(δ j p t+1 ) = log p t+1(j) p t (j) = log 1 Z t+1 ηl t (j) ψ t := log E I pt exp( η(l t (I ) E I p t l t (I ))) = ηel t (I ) + log(z t+1 )
31 Adversarial multi-armed bandit, fundamental strategy Exponential weights strategy for full information (l t is observed at the end of round t): play I t at random from p t where p t+1 (i) = 1 Z t+1 p t (i) exp( ηl t (i)). In five lines one can show R T 2T log(n) with p 1 (i) = 1/n: Ent(δ j p t ) Ent(δ j p t+1 ) = log p t+1(j) p t (j) = log 1 Z t+1 ηl t (j) ψ t := log E I pt exp( η(l t (I ) E I p t l t (I ))) = ηel t (I ) + log(z t+1 ) η ( ) p t (i)l t (i) l t (j) = Ent(δ j p 1 ) Ent(δ j p T +1 ) + t t i ψ t
32 Adversarial multi-armed bandit, fundamental strategy Exponential weights strategy for full information (l t is observed at the end of round t): play I t at random from p t where p t+1 (i) = 1 Z t+1 p t (i) exp( ηl t (i)). In five lines one can show R T 2T log(n) with p 1 (i) = 1/n: Ent(δ j p t ) Ent(δ j p t+1 ) = log p t+1(j) p t (j) = log 1 Z t+1 ηl t (j) ψ t := log E I pt exp( η(l t (I ) E I p t l t (I ))) = ηel t (I ) + log(z t+1 ) η ( ) p t (i)l t (i) l t (j) = Ent(δ j p 1 ) Ent(δ j p T +1 ) + t t i Using that l t 0 one has ψ t η2 2 El t(i) 2 thus R T log(n) + ηt η 2 ψ t
33 Adversarial multi-armed bandit, fundamental strategy Exp3: replace l t by l t in the exponential weights strategy, where l t (i) = l t(i t ) p t (i) 1{i = I t}. Key property: E It p t lt (i) = l t (i).
34 Adversarial multi-armed bandit, fundamental strategy Exp3: replace l t by l t in the exponential weights strategy, where l t (i) = l t(i t ) p t (i) 1{i = I t}. Key property: E It p t lt (i) = l t (i). Thus with the analysis from the previous slide: R T log(n) η + η 2 E t E I pt lt (I ) 2.
35 Adversarial multi-armed bandit, fundamental strategy Exp3: replace l t by l t in the exponential weights strategy, where l t (i) = l t(i t ) p t (i) 1{i = I t}. Key property: E It p t lt (i) = l t (i). Thus with the analysis from the previous slide: R T log(n) η + η 2 E t E I pt lt (I ) 2. Amazingly the variance term is automatically controlled: E It,I p t lt (I ) 2 E It,I p t 1{I = I t } p t (I t ) 2 = E I pt 1 p t (I ) = n.
36 Adversarial multi-armed bandit, fundamental strategy Exp3: replace l t by l t in the exponential weights strategy, where l t (i) = l t(i t ) p t (i) 1{i = I t}. Key property: E It p t lt (i) = l t (i). Thus with the analysis from the previous slide: R T log(n) η + η 2 E t E I pt lt (I ) 2. Amazingly the variance term is automatically controlled: E It,I p t lt (I ) 2 E It,I p t 1{I = I t } p t (I t ) 2 = E I pt 1 p t (I ) = n. Thus with η = 2n log(n)/t one gets R T 2Tn log(n).
37 Adversarial multi-armed bandit, going further 1. With the modified loss estimate lt(it)1{i=it}+β p t(i t) one can prove high probability bounds on R T, and by integrating the deviations one can show ER T = O( Tn log(n)).
38 Adversarial multi-armed bandit, going further 1. With the modified loss estimate lt(it)1{i=it}+β p t(i t) one can prove high probability bounds on R T, and by integrating the deviations one can show ER T = O( Tn log(n)). 2. The extraneous logarithmic factor in the pseudo-regret upper can be removed, see Audibert and Bubeck [2009]. Conjecture: one cannot remove the log factor for the expected regret, that is for any strategy there exists an adaptive adversary such that ER T = Ω( Tn log(n)).
39 Adversarial multi-armed bandit, going further 1. With the modified loss estimate lt(it)1{i=it}+β p t(i t) one can prove high probability bounds on R T, and by integrating the deviations one can show ER T = O( Tn log(n)). 2. The extraneous logarithmic factor in the pseudo-regret upper can be removed, see Audibert and Bubeck [2009]. Conjecture: one cannot remove the log factor for the expected regret, that is for any strategy there exists an adaptive adversary such that ER T = Ω( Tn log(n)). 3. T can be replaced by various measure of variance in the loss sequence, see e.g., Hazan and Kale [2009].
40 Adversarial multi-armed bandit, going further 1. With the modified loss estimate lt(it)1{i=it}+β p t(i t) one can prove high probability bounds on R T, and by integrating the deviations one can show ER T = O( Tn log(n)). 2. The extraneous logarithmic factor in the pseudo-regret upper can be removed, see Audibert and Bubeck [2009]. Conjecture: one cannot remove the log factor for the expected regret, that is for any strategy there exists an adaptive adversary such that ER T = Ω( Tn log(n)). 3. T can be replaced by various measure of variance in the loss sequence, see e.g., Hazan and Kale [2009]. 4. There exists strategies which guarantee simultaneously R T = Õ( Tn) in the adversarial model and R T = Õ( i 1 i ) in the i.i.d. model, see Bubeck and Slivkins [2012].
41 Adversarial multi-armed bandit, going further 1. With the modified loss estimate lt(it)1{i=it}+β p t(i t) one can prove high probability bounds on R T, and by integrating the deviations one can show ER T = O( Tn log(n)). 2. The extraneous logarithmic factor in the pseudo-regret upper can be removed, see Audibert and Bubeck [2009]. Conjecture: one cannot remove the log factor for the expected regret, that is for any strategy there exists an adaptive adversary such that ER T = Ω( Tn log(n)). 3. T can be replaced by various measure of variance in the loss sequence, see e.g., Hazan and Kale [2009]. 4. There exists strategies which guarantee simultaneously R T = Õ( Tn) in the adversarial model and R T = Õ( i 1 i ) in the i.i.d. model, see Bubeck and Slivkins [2012]. 5. Graph feedback structure, regret with respect to S switches, label efficient, switching cost...
42 Bayesian multi-armed bandit, Thompson [1933] Set of models {(ν 1 (θ),..., ν n (θ)), θ Θ} and prior distribution π 0 over Θ. The Bayesian regret is defined as BR T (π 0 ) = E θ π0 R T (ν 1 (θ),..., ν n (θ)).
43 Bayesian multi-armed bandit, Thompson [1933] Set of models {(ν 1 (θ),..., ν n (θ)), θ Θ} and prior distribution π 0 over Θ. The Bayesian regret is defined as BR T (π 0 ) = E θ π0 R T (ν 1 (θ),..., ν n (θ)). In principle the strategy minimizing the Bayesian regret can be computed by dynamic programming on the potentially huge state space P(Θ).
44 Bayesian multi-armed bandit, Thompson [1933] Set of models {(ν 1 (θ),..., ν n (θ)), θ Θ} and prior distribution π 0 over Θ. The Bayesian regret is defined as BR T (π 0 ) = E θ π0 R T (ν 1 (θ),..., ν n (θ)). In principle the strategy minimizing the Bayesian regret can be computed by dynamic programming on the potentially huge state space P(Θ). The celebrated Gittins index theorem gives sufficient condition to dramatically reduce the computational complexity of implementing the optimal Bayesian strategy under a strong product assumption on π 0.
45 Bayesian multi-armed bandit, Thompson [1933] Set of models {(ν 1 (θ),..., ν n (θ)), θ Θ} and prior distribution π 0 over Θ. The Bayesian regret is defined as BR T (π 0 ) = E θ π0 R T (ν 1 (θ),..., ν n (θ)). In principle the strategy minimizing the Bayesian regret can be computed by dynamic programming on the potentially huge state space P(Θ). The celebrated Gittins index theorem gives sufficient condition to dramatically reduce the computational complexity of implementing the optimal Bayesian strategy under a strong product assumption on π 0. Notation: π t denotes the posterior distribution on θ at time t.
46 Bayesian multi-armed bandit, Gittins index Theorem (Gittins [1979]) Consider the product and γ-discounted case: Θ = i Θ i, ν i (θ) := ν(θ i ), π 0 = i π 0 (i), and furthermore one is interested in maximizing E t 0 γt Y t.
47 Bayesian multi-armed bandit, Gittins index Theorem (Gittins [1979]) Consider the product and γ-discounted case: Θ = i Θ i, ν i (θ) := ν(θ i ), π 0 = i π 0 (i), and furthermore one is interested in maximizing E t 0 γt Y t. The optimal Bayesian strategy is to pick at time s the arm maximizing: sup { λ R : sup E τ ( γ t X t + t<τ ) } γτ 1 γ λ 1 1 γ λ, where the expectation is over (X t ) drawn from ν(θ) with θ π s (i), and the supremum is taken over all stopping times τ.
48 Bayesian multi-armed bandit, Gittins index Theorem (Gittins [1979]) Consider the product and γ-discounted case: Θ = i Θ i, ν i (θ) := ν(θ i ), π 0 = i π 0 (i), and furthermore one is interested in maximizing E t 0 γt Y t. The optimal Bayesian strategy is to pick at time s the arm maximizing: sup { λ R : sup E τ ( γ t X t + t<τ ) } γτ 1 γ λ 1 1 γ λ, where the expectation is over (X t ) drawn from ν(θ) with θ π s (i), and the supremum is taken over all stopping times τ. For much more (implementation for exponential families, interpretation as a multitoken Markov game,...) see Dumitriu, Tetali and Winkler [2003], Gittins, Glazebrook, Weber [2011], Kaufmann [2014].
49 Bayesian multi-armed bandit, Gittins index Weber [1992] gives an{ exquisite proof of Gittins theorem. } Let λ t (i) := sup λ R : sup E γ t (X t λ) 0 τ t<τ the Gittins index of arm i at time t, which we interpret as the maximum charge one is willing to pay to play arm i given the current information.
50 Bayesian multi-armed bandit, Gittins index Weber [1992] gives an{ exquisite proof of Gittins theorem. } Let λ t (i) := sup λ R : sup E γ t (X t λ) 0 τ t<τ the Gittins index of arm i at time t, which we interpret as the maximum charge one is willing to pay to play arm i given the current information. The prevailing charge is defined as min s t λ s (i) (i.e. whenever the prevailing charge is too high we just drop it to the fair level).
51 Bayesian multi-armed bandit, Gittins index Weber [1992] gives an{ exquisite proof of Gittins theorem. } Let λ t (i) := sup λ R : sup E γ t (X t λ) 0 τ t<τ the Gittins index of arm i at time t, which we interpret as the maximum charge one is willing to pay to play arm i given the current information. The prevailing charge is defined as min s t λ s (i) (i.e. whenever the prevailing charge is too high we just drop it to the fair level). 1. Discounted sum of prevailing charge for played arms is an upper bound (in expectation) on the discounted sum of rewards.
52 Bayesian multi-armed bandit, Gittins index Weber [1992] gives an{ exquisite proof of Gittins theorem. } Let λ t (i) := sup λ R : sup E γ t (X t λ) 0 τ t<τ the Gittins index of arm i at time t, which we interpret as the maximum charge one is willing to pay to play arm i given the current information. The prevailing charge is defined as min s t λ s (i) (i.e. whenever the prevailing charge is too high we just drop it to the fair level). 1. Discounted sum of prevailing charge for played arms is an upper bound (in expectation) on the discounted sum of rewards. 2. Since the prevailing charge is nonincreasing, the discounted sum of prevailing charge is maximized if we always pick the arm with maximum prevailing charge.
53 Bayesian multi-armed bandit, Gittins index Weber [1992] gives an{ exquisite proof of Gittins theorem. } Let λ t (i) := sup λ R : sup E γ t (X t λ) 0 τ t<τ the Gittins index of arm i at time t, which we interpret as the maximum charge one is willing to pay to play arm i given the current information. The prevailing charge is defined as min s t λ s (i) (i.e. whenever the prevailing charge is too high we just drop it to the fair level). 1. Discounted sum of prevailing charge for played arms is an upper bound (in expectation) on the discounted sum of rewards. 2. Since the prevailing charge is nonincreasing, the discounted sum of prevailing charge is maximized if we always pick the arm with maximum prevailing charge. 3. Gittins index does exactly 2. and that in this case 1. is an equality. Q.E.D.
54 Bayesian multi-armed bandit, Thompson Sampling (TS) In machine learning we want (i) strategies that can deal with complicated priors, and (ii) guarantees for misspecified priors. This is why we have to go beyond the Gittins index theory.
55 Bayesian multi-armed bandit, Thompson Sampling (TS) In machine learning we want (i) strategies that can deal with complicated priors, and (ii) guarantees for misspecified priors. This is why we have to go beyond the Gittins index theory. In his 1933 paper Thompson proposed the following strategy: sample θ π t and play I t argmax µ i (θ ).
56 Bayesian multi-armed bandit, Thompson Sampling (TS) In machine learning we want (i) strategies that can deal with complicated priors, and (ii) guarantees for misspecified priors. This is why we have to go beyond the Gittins index theory. In his 1933 paper Thompson proposed the following strategy: sample θ π t and play I t argmax µ i (θ ). Theoretical guarantees for this highly practical strategy have long remained elusive. Recently Agrawal and Goyal [2012] and Kaufmann, Korda and Munos [2012] proved that TS with Bernoulli reward distributions ( and uniform prior on the parameters achieves ) log(t ) R T = O i i (note that this is the frequentist regret!).
57 Bayesian multi-armed bandit, Thompson Sampling (TS) In machine learning we want (i) strategies that can deal with complicated priors, and (ii) guarantees for misspecified priors. This is why we have to go beyond the Gittins index theory. In his 1933 paper Thompson proposed the following strategy: sample θ π t and play I t argmax µ i (θ ). Theoretical guarantees for this highly practical strategy have long remained elusive. Recently Agrawal and Goyal [2012] and Kaufmann, Korda and Munos [2012] proved that TS with Bernoulli reward distributions ( and uniform prior on the parameters achieves ) log(t ) R T = O i i (note that this is the frequentist regret!). Guha and Munagala [2014] conjecture that, for product priors, TS is a 2-approximation to the optimal Bayesian strategy for the objective of minimizing the number of pulls on suboptimal arms.
58 Bayesian multi-armed bandit, Russo and Van Roy [2014] information ratio analysis Assume a prior in the adversarial model, that is a prior over (l 1,..., l T ) [0, 1] n T, and let E t denote the posterior distribution (given l 1 (I 1 ),..., l t 1 (I t 1 )).
59 Bayesian multi-armed bandit, Russo and Van Roy [2014] information ratio analysis Assume a prior in the adversarial model, that is a prior over (l 1,..., l T ) [0, 1] n T, and let E t denote the posterior distribution (given l 1 (I 1 ),..., l t 1 (I t 1 )). We introduce r t (i) = E t (l t (i) l t (i )), and v t (i) = Var t (E t (l t (i) i )).
60 Bayesian multi-armed bandit, Russo and Van Roy [2014] information ratio analysis Assume a prior in the adversarial model, that is a prior over (l 1,..., l T ) [0, 1] n T, and let E t denote the posterior distribution (given l 1 (I 1 ),..., l t 1 (I t 1 )). We introduce r t (i) = E t (l t (i) l t (i )), and v t (i) = Var t (E t (l t (i) i )). Key observation (next slide): E v t (I t ) 1 2 H(i ) t T
61 Bayesian multi-armed bandit, Russo and Van Roy [2014] information ratio analysis Assume a prior in the adversarial model, that is a prior over (l 1,..., l T ) [0, 1] n T, and let E t denote the posterior distribution (given l 1 (I 1 ),..., l t 1 (I t 1 )). We introduce r t (i) = E t (l t (i) l t (i )), and v t (i) = Var t (E t (l t (i) i )). Key observation (next slide): which implies: E t T v t (I t ) 1 2 H(i ) t, E t r t (I t ) C E t v t (I t ) T T E r t (I t ) C Evt (I t ) t=1 t=1 BR T C T H(i )/2.
62 Bayesian multi-armed bandit, accumulation of information v t (i) = Var t (E t (l t (i) i )), π t (j) = P t (i = j), E t T v t (I t ) 1 2 H(x )
63 Bayesian multi-armed bandit, accumulation of information v t (i) = Var t (E t (l t (i) i )), π t (j) = P t (i = j), E t T v t (I t ) 1 2 H(x ) Equipped with Pinsker s inequality and basic information theory concepts (such as the mutual information I) one has: v t (i) = j π t (j)(e t (l t (i) i = j) E t (l t (i))) 2 1 π t (j)ent(l t (l t (i) i = j) L t (l t (i))) 2 j = 1 2 I t(l t (i), i ) = H t (i ) H t (i l t (i)).
64 Bayesian multi-armed bandit, accumulation of information v t (i) = Var t (E t (l t (i) i )), π t (j) = P t (i = j), E t T v t (I t ) 1 2 H(x ) Equipped with Pinsker s inequality and basic information theory concepts (such as the mutual information I) one has: v t (i) = j π t (j)(e t (l t (i) i = j) E t (l t (i))) 2 1 π t (j)ent(l t (l t (i) i = j) L t (l t (i))) 2 j = 1 2 I t(l t (i), i ) = H t (i ) H t (i l t (i)). Thus Ev t (I t ) 1 2 E(H t(i ) H t+1 (i )).
65 Bayesian multi-armed bandit, TS information ratio Let l t (i) = E t l t (i) and l t (i, j) = E t (l t (i) i = j). Then E t r t (I t ) C E t v t (I t ) E t l t (I t ) π t (i) l t (i, i) C E t π t (j)( l t (I t, j) l t (I t )) 2 i j
66 Bayesian multi-armed bandit, TS information ratio Let l t (i) = E t l t (i) and l t (i, j) = E t (l t (i) i = j). Then E t r t (I t ) C E t v t (I t ) E t l t (I t ) π t (i) l t (i, i) C E t π t (j)( l t (I t, j) l t (I t )) 2 i j For TS the following shows that one can take C = n: E t l t (I t ) i π t (i) l t (i, i) = i n i π t (i)( l t (i) l t (i, i)) π t (i) 2 ( l t (i) l t (i, i)) 2 n i,j π t (i)π t (j)( l t (i) l t (i, j)) 2. Thus TS always satisfies BR T TnH(i ) Tn log(n).
67 Bayesian multi-armed bandit, TS information ratio Let l t (i) = E t l t (i) and l t (i, j) = E t (l t (i) i = j). Then E t r t (I t ) C E t v t (I t ) E t l t (I t ) π t (i) l t (i, i) C E t π t (j)( l t (I t, j) l t (I t )) 2 i j For TS the following shows that one can take C = n: E t l t (I t ) i π t (i) l t (i, i) = i n i π t (i)( l t (i) l t (i, i)) π t (i) 2 ( l t (i) l t (i, i)) 2 n i,j π t (i)π t (j)( l t (i) l t (i, j)) 2. Thus TS always satisfies BR T TnH(i ) Tn log(n). Side note: by the minimax theorem this implies there exists a strategy for the oblivious adversarial model with regret Tn log(n).
68 Summary of basic results ) log(t ) i ( 1. In the i.i.d. model UCB attains a regret of O i by Lai and Robbins lower bound this is optimal (up to a multiplicative variance term). and 2. In the adversarial model Exp3 attains a regret of O( Tn log(n)) and this is optimal up to the logarithmic term. 3. In the Bayesian model, Gittins index gives an optimal strategy for the case of product priors. For general priors Thompson Sampling is a more flexible strategy. Its Bayesian regret is controlled by the entropy of the optimal decision. Moreover TS with an uninformative prior has frequentist guarantees comparable to UCB.
69 Part 2: Linear, non-linear, and contextual bandit
70 The linear bandit problem, Auer [2002] Known parameters: compact action set A R n, adversary s action set L R n, number of rounds T.
71 The linear bandit problem, Auer [2002] Known parameters: compact action set A R n, adversary s action set L R n, number of rounds T. Protocol: For each round t = 1, 2,..., T, the adversary chooses a loss vector l t L and simultaneously the player chooses a t A based on past observations and receives a loss/observation Y t = l t a t. R T = E T l t a t min E T l t a. a A t=1 t=1
72 The linear bandit problem, Auer [2002] Known parameters: compact action set A R n, adversary s action set L R n, number of rounds T. Protocol: For each round t = 1, 2,..., T, the adversary chooses a loss vector l t L and simultaneously the player chooses a t A based on past observations and receives a loss/observation Y t = l t a t. R T = E T l t a t min E T l t a. a A t=1 Other models: In the i.i.d. model we assume that there is some underlying θ L such that E(Y t a t ) = θ a t. In the Bayesian model we assume that we have a prior distribution ν over the sequence (l 1,..., l T ) (in this case the expectation in R T is also over (l 1,..., l T ) ν). Alternatively we could assume a prior over θ. t=1
73 The linear bandit problem, Auer [2002] Known parameters: compact action set A R n, adversary s action set L R n, number of rounds T. Protocol: For each round t = 1, 2,..., T, the adversary chooses a loss vector l t L and simultaneously the player chooses a t A based on past observations and receives a loss/observation Y t = l t a t. R T = E T l t a t min E T l t a. a A t=1 Other models: In the i.i.d. model we assume that there is some underlying θ L such that E(Y t a t ) = θ a t. In the Bayesian model we assume that we have a prior distribution ν over the sequence (l 1,..., l T ) (in this case the expectation in R T is also over (l 1,..., l T ) ν). Alternatively we could assume a prior over θ. Example: Part 1 was about A = {e 1,..., e n } and L = [0, 1] n. t=1
74 The linear bandit problem, Auer [2002] Known parameters: compact action set A R n, adversary s action set L R n, number of rounds T. Protocol: For each round t = 1, 2,..., T, the adversary chooses a loss vector l t L and simultaneously the player chooses a t A based on past observations and receives a loss/observation Y t = l t a t. R T = E T l t a t min E T l t a. a A t=1 Other models: In the i.i.d. model we assume that there is some underlying θ L such that E(Y t a t ) = θ a t. In the Bayesian model we assume that we have a prior distribution ν over the sequence (l 1,..., l T ) (in this case the expectation in R T is also over (l 1,..., l T ) ν). Alternatively we could assume a prior over θ. Example: Part 1 was about A = {e 1,..., e n } and L = [0, 1] n. Assumption: unless specified otherwise we assume L = A := {l : sup a A l a 1}. t=1
75 Example: path planning
76 Example: path planning Adversary Player
77 Example: path planning Adversary Player
78 Example: path planning Adversary Player
79 Example: path planning Adversary l 1 l 4 l 5 l n 2 l 2 l 6 l n 1 l 3 l 8 l 7 l n l 9 Player
80 Example: path planning Adversary l 1 l 4 l 5 l n 2 l 2 l 6 l n 1 l 3 l 8 l 7 l n l 9 Player loss suffered: l 2 + l l n
81 Example: path planning Adversary Feedback: Full Info: l 1, l 2,..., l n l 1 l 4 l 5 l n 2 l 2 l 6 l n 1 l 3 l 8 l 7 l n l 9 Player loss suffered: l 2 + l l n
82 Example: path planning Adversary Feedback: Full Info: Semi-Bandit: l 1, l 2,..., l n l 2, l 7,..., l n l 1 l 4 l 5 l n 2 l 2 l 6 l n 1 l 3 l 8 l 7 l n l 9 Player loss suffered: l 2 + l l n
83 Example: path planning Adversary Feedback: Full Info: Semi-Bandit: Bandit: l 1, l 2,..., l n l 2, l 7,..., l n l 2 + l l n l 1 l 4 l 5 l n 2 l 2 l 6 l n 1 l 3 l 8 l 7 l n l 9 Player loss suffered: l 2 + l l n
84 Thompson Sampling for linear bandit after RVR14 Assume A = {a 1,..., a A }. Recall from Part 1 that TS satisfies π t (i)( l t (i) l t (i, i)) C π t (i)π t (j)( l t (i, j) l t (i)) 2 i i,j R T C T log( A )/2, where l t (i) = E t l t (i) and l t (i, j) = E t (l t (i) i = j).
85 Thompson Sampling for linear bandit after RVR14 Assume A = {a 1,..., a A }. Recall from Part 1 that TS satisfies π t (i)( l t (i) l t (i, i)) C π t (i)π t (j)( l t (i, j) l t (i)) 2 i i,j R T C T log( A )/2, where l t (i) = E t l t (i) and l t (i, j) = E t (l t (i) i = j). Writing l ( t (i) = ai l t, l t (i, j) = ai l j t, and πt ) (M i,j ) = (i)π t (j)ai ( l t l j t) we want to show that Tr(M) C M F.
86 Thompson Sampling for linear bandit after RVR14 Assume A = {a 1,..., a A }. Recall from Part 1 that TS satisfies π t (i)( l t (i) l t (i, i)) C π t (i)π t (j)( l t (i, j) l t (i)) 2 i i,j R T C T log( A )/2, where l t (i) = E t l t (i) and l t (i, j) = E t (l t (i) i = j). Writing l ( t (i) = ai l t, l t (i, j) = ai l j t, and πt ) (M i,j ) = (i)π t (j)ai ( l t l j t) we want to show that Tr(M) C M F. Using the eigenvalue formula for the trace and the Frobenius norm one can see that Tr(M) 2 rank(m) M 2 F.
87 Thompson Sampling for linear bandit after RVR14 Assume A = {a 1,..., a A }. Recall from Part 1 that TS satisfies π t (i)( l t (i) l t (i, i)) C π t (i)π t (j)( l t (i, j) l t (i)) 2 i i,j R T C T log( A )/2, where l t (i) = E t l t (i) and l t (i, j) = E t (l t (i) i = j). Writing l ( t (i) = ai l t, l t (i, j) = ai l j t, and πt ) (M i,j ) = (i)π t (j)ai ( l t l j t) we want to show that Tr(M) C M F. Using the eigenvalue formula for the trace and the Frobenius norm one can see that Tr(M) 2 rank(m) M 2 F. Moreover the rank of M is at most n since M = UV where U, V R A n (the i th row of U is π t (i)a i and for V it is π t (i)( l t l i t)).
88 Thompson Sampling for linear bandit after RVR14 1. TS satisfies R T nt log( A ). To appreciate the improvement recall that without the linear structure one would get a regret of order A T and that A can be exponential in the dimension n (think of the path planning example).
89 Thompson Sampling for linear bandit after RVR14 1. TS satisfies R T nt log( A ). To appreciate the improvement recall that without the linear structure one would get a regret of order A T and that A can be exponential in the dimension n (think of the path planning example). 2. Provided that one can efficiently sample from the posterior on l t (or on θ), TS just requires at each step one linear optimization over A.
90 Thompson Sampling for linear bandit after RVR14 1. TS satisfies R T nt log( A ). To appreciate the improvement recall that without the linear structure one would get a regret of order A T and that A can be exponential in the dimension n (think of the path planning example). 2. Provided that one can efficiently sample from the posterior on l t (or on θ), TS just requires at each step one linear optimization over A. 3. TS regret bound is optimal in the following sense. W.l.og. one can assume A (10T ) n and thus TS satisfies R T = O(n T log(t )) for any action set. Furthermore one can show that there exists an action set and a prior such that for any strategy one has R T = Ω(n T ), see Dani, Hayes and Kakade [2008], Rusmevichientong and Tsitsiklis [2010], and Audibert, Bubeck and Lugosi [2011, 2014].
91 Adversarial linear bandit after Dani, Hayes, Kakade [2008] Recall from Part 1 that exponential weights satisfies for any l t such that E l t (i) = l t (i) and l t (i) 0, R T max i Ent(δ i p 1 ) η + η 2 E t E I pt lt (I ) 2.
92 Adversarial linear bandit after Dani, Hayes, Kakade [2008] Recall from Part 1 that exponential weights satisfies for any l t such that E l t (i) = l t (i) and l t (i) 0, R T max i Ent(δ i p 1 ) η + η 2 E t E I pt lt (I ) 2. DHK08 proposed the following (beautiful) unbiased estimator for the linear case: l t = Σ 1 t a t a t l t where Σ t = E a pt (aa ).
93 Adversarial linear bandit after Dani, Hayes, Kakade [2008] Recall from Part 1 that exponential weights satisfies for any l t such that E l t (i) = l t (i) and l t (i) 0, R T max i Ent(δ i p 1 ) η + η 2 E t E I pt lt (I ) 2. DHK08 proposed the following (beautiful) unbiased estimator for the linear case: l t = Σ 1 t a t a t l t where Σ t = E a pt (aa ). Again, amazingly, the variance is automatically controlled: E(E a pt ( l t a) 2 ) = E l t Σ t lt Eat Σ 1 t a t = ETr(Σ 1 t a t a t ) = n.
94 Adversarial linear bandit after Dani, Hayes, Kakade [2008] Recall from Part 1 that exponential weights satisfies for any l t such that E l t (i) = l t (i) and l t (i) 0, R T max i Ent(δ i p 1 ) η + η 2 E t E I pt lt (I ) 2. DHK08 proposed the following (beautiful) unbiased estimator for the linear case: l t = Σ 1 t a t a t l t where Σ t = E a pt (aa ). Again, amazingly, the variance is automatically controlled: E(E a pt ( l t a) 2 ) = E l t Σ t lt Eat Σ 1 t a t = ETr(Σ 1 t a t a t ) = n. Up to the issue that l t can take negative values this suggests the optimal nt log( A ) regret bound.
95 Adversarial linear bandit, further development 1. The non-negativity issue of l t is a manifestation of the need for an added exploration. DHK08 used a suboptimal exploration which led to an additional n in the regret. This was later improved in Bubeck, Cesa-Bianchi, and Kakade [2012] with an exploration based on the John s ellipsoid (smallest ellipsoid containing A).
96 Adversarial linear bandit, further development 1. The non-negativity issue of l t is a manifestation of the need for an added exploration. DHK08 used a suboptimal exploration which led to an additional n in the regret. This was later improved in Bubeck, Cesa-Bianchi, and Kakade [2012] with an exploration based on the John s ellipsoid (smallest ellipsoid containing A). 2. Sampling the exp. weights is usually computationally difficult, see Cesa-Bianchi and Lugosi [2009] for some exceptions.
97 Adversarial linear bandit, further development 1. The non-negativity issue of l t is a manifestation of the need for an added exploration. DHK08 used a suboptimal exploration which led to an additional n in the regret. This was later improved in Bubeck, Cesa-Bianchi, and Kakade [2012] with an exploration based on the John s ellipsoid (smallest ellipsoid containing A). 2. Sampling the exp. weights is usually computationally difficult, see Cesa-Bianchi and Lugosi [2009] for some exceptions. 3. Abernethy, Hazan and Rakhlin [2008] proposed an alternative (beautiful) strategy based on mirror descent. The key idea is to use a n-self-concordant barrier for conv(a) as a mirror map and to sample points uniformly in Dikin ellipses. This method s regret is suboptimal by a factor n and the computational efficiency depends on the barrier being used.
98 Adversarial linear bandit, further development 1. The non-negativity issue of l t is a manifestation of the need for an added exploration. DHK08 used a suboptimal exploration which led to an additional n in the regret. This was later improved in Bubeck, Cesa-Bianchi, and Kakade [2012] with an exploration based on the John s ellipsoid (smallest ellipsoid containing A). 2. Sampling the exp. weights is usually computationally difficult, see Cesa-Bianchi and Lugosi [2009] for some exceptions. 3. Abernethy, Hazan and Rakhlin [2008] proposed an alternative (beautiful) strategy based on mirror descent. The key idea is to use a n-self-concordant barrier for conv(a) as a mirror map and to sample points uniformly in Dikin ellipses. This method s regret is suboptimal by a factor n and the computational efficiency depends on the barrier being used. 4. Bubeck and Eldan [2014] s entropic barrier allows for a much more information-efficient sampling than AHR08. This gives another strategy with optimal regret which is efficient when A is convex (and one can do linear optimization on A).
99 Adversarial combinatorial bandit after Audibert, Bubeck and Lugosi [2011, 2014] Combinatorial setting: A {0, 1} n, max a a 1 = m, L = [0, 1] n.
100 Adversarial combinatorial bandit after Audibert, Bubeck and Lugosi [2011, 2014] Combinatorial setting: A {0, 1} n, max a a 1 = m, L = [0, 1] n. 1. Full information case goes back to the end of the 90 s (Warmuth and co-authors), semi-bandit and bandit were introduced in Audibert, Bubeck and Lugosi [2011] (following several papers that studied specific sets A).
101 Adversarial combinatorial bandit after Audibert, Bubeck and Lugosi [2011, 2014] Combinatorial setting: A {0, 1} n, max a a 1 = m, L = [0, 1] n. 1. Full information case goes back to the end of the 90 s (Warmuth and co-authors), semi-bandit and bandit were introduced in Audibert, Bubeck and Lugosi [2011] (following several papers that studied specific sets A). 2. This is a natural setting to study FPL-type (Follow the Perturbed Leader) strategies, see e.g. Kalai and Vempala [2004] and more recently Devroye, Lugosi and Neu [2013].
102 Adversarial combinatorial bandit after Audibert, Bubeck and Lugosi [2011, 2014] Combinatorial setting: A {0, 1} n, max a a 1 = m, L = [0, 1] n. 1. Full information case goes back to the end of the 90 s (Warmuth and co-authors), semi-bandit and bandit were introduced in Audibert, Bubeck and Lugosi [2011] (following several papers that studied specific sets A). 2. This is a natural setting to study FPL-type (Follow the Perturbed Leader) strategies, see e.g. Kalai and Vempala [2004] and more recently Devroye, Lugosi and Neu [2013]. 3. ABL11: Exponential weights is provably suboptimal in this setting! This is in sharp contrast with the case where L = A.
103 Adversarial combinatorial bandit after Audibert, Bubeck and Lugosi [2011, 2014] Combinatorial setting: A {0, 1} n, max a a 1 = m, L = [0, 1] n. 1. Full information case goes back to the end of the 90 s (Warmuth and co-authors), semi-bandit and bandit were introduced in Audibert, Bubeck and Lugosi [2011] (following several papers that studied specific sets A). 2. This is a natural setting to study FPL-type (Follow the Perturbed Leader) strategies, see e.g. Kalai and Vempala [2004] and more recently Devroye, Lugosi and Neu [2013]. 3. ABL11: Exponential weights is provably suboptimal in this setting! This is in sharp contrast with the case where L = A. 4. Optimal regret in the semi-bandit case is mnt and it can be achieved with mirror descent and the natural unbiased estimator for the semi-bandit situation.
104 Adversarial combinatorial bandit after Audibert, Bubeck and Lugosi [2011, 2014] Combinatorial setting: A {0, 1} n, max a a 1 = m, L = [0, 1] n. 1. Full information case goes back to the end of the 90 s (Warmuth and co-authors), semi-bandit and bandit were introduced in Audibert, Bubeck and Lugosi [2011] (following several papers that studied specific sets A). 2. This is a natural setting to study FPL-type (Follow the Perturbed Leader) strategies, see e.g. Kalai and Vempala [2004] and more recently Devroye, Lugosi and Neu [2013]. 3. ABL11: Exponential weights is provably suboptimal in this setting! This is in sharp contrast with the case where L = A. 4. Optimal regret in the semi-bandit case is mnt and it can be achieved with mirror descent and the natural unbiased estimator for the semi-bandit situation. 5. For the bandit case the bound for exponential weights from the previous slides gives m mnt. However the lower bound from ABL14 is m nt, which is conjectured to be tight.
105 Preliminaries for the i.i.d. case: a primer on least squares Assume Y t = θ a t + ξ t where (ξ t ) is an i.i.d. sequence of centered and sub-gaussian real-valued random variables. The (regularized) least squares estimator for θ based on Y t = (Y 1,..., Y t 1 ) is, with A t = (a 1... a t 1 ) R n t 1 and Σ t = λi n + t 1 s=1 a sa s : ˆθ t = Σ 1 t A t Y t
106 Preliminaries for the i.i.d. case: a primer on least squares Assume Y t = θ a t + ξ t where (ξ t ) is an i.i.d. sequence of centered and sub-gaussian real-valued random variables. The (regularized) least squares estimator for θ based on Y t = (Y 1,..., Y t 1 ) is, with A t = (a 1... a t 1 ) R n t 1 and Σ t = λi n + t 1 s=1 a sa s : ˆθ t = Σ 1 t A t Y t Observe that we can also write θ = Σ 1 t (A t (Y t + ε t ) + λθ) where ε t = (E(Y 1 a 1 ) Y 1,..., E(Y t 1 a t 1 ) Y t 1 )
107 Preliminaries for the i.i.d. case: a primer on least squares Assume Y t = θ a t + ξ t where (ξ t ) is an i.i.d. sequence of centered and sub-gaussian real-valued random variables. The (regularized) least squares estimator for θ based on Y t = (Y 1,..., Y t 1 ) is, with A t = (a 1... a t 1 ) R n t 1 and Σ t = λi n + t 1 s=1 a sa s : ˆθ t = Σ 1 t A t Y t Observe that we can also write θ = Σ 1 t (A t (Y t + ε t ) + λθ) where ε t = (E(Y 1 a 1 ) Y 1,..., E(Y t 1 a t 1 ) Y t 1 ) so that θ ˆθ t Σt = A t ε t + λθ Σ 1 t A t ε t Σ 1 t + λ θ.
108 Preliminaries for the i.i.d. case: a primer on least squares Assume Y t = θ a t + ξ t where (ξ t ) is an i.i.d. sequence of centered and sub-gaussian real-valued random variables. The (regularized) least squares estimator for θ based on Y t = (Y 1,..., Y t 1 ) is, with A t = (a 1... a t 1 ) R n t 1 and Σ t = λi n + t 1 s=1 a sa s : ˆθ t = Σ 1 t A t Y t Observe that we can also write θ = Σ 1 t (A t (Y t + ε t ) + λθ) where ε t = (E(Y 1 a 1 ) Y 1,..., E(Y t 1 a t 1 ) Y t 1 ) so that θ ˆθ t Σt = A t ε t + λθ Σ 1 t A t ε t Σ 1 t + λ θ. A basic martingale argument (see e.g., Abbasi-Yadkori, Pál and Szepesvári [2011]) shows that w.p. 1 δ, t 1, A t ε t Σ 1 t logdet(σ t ) + log(1/(δ 2 λ n )).
109 Preliminaries for the i.i.d. case: a primer on least squares Assume Y t = θ a t + ξ t where (ξ t ) is an i.i.d. sequence of centered and sub-gaussian real-valued random variables. The (regularized) least squares estimator for θ based on Y t = (Y 1,..., Y t 1 ) is, with A t = (a 1... a t 1 ) R n t 1 and Σ t = λi n + t 1 s=1 a sa s : ˆθ t = Σ 1 t A t Y t Observe that we can also write θ = Σ 1 t (A t (Y t + ε t ) + λθ) where ε t = (E(Y 1 a 1 ) Y 1,..., E(Y t 1 a t 1 ) Y t 1 ) so that θ ˆθ t Σt = A t ε t + λθ Σ 1 t A t ε t Σ 1 t + λ θ. A basic martingale argument (see e.g., Abbasi-Yadkori, Pál and Szepesvári [2011]) shows that w.p. 1 δ, t 1, A t ε t Σ 1 t logdet(σ t ) + log(1/(δ 2 λ n )). Note that logdet(σ t ) n log(tr(σ t )/n) n log(λ + t/n) (w.l.o.g. we assumed a t 1).
110 i.i.d. linear bandit after DHK08, RT10, AYPS11 Let β = 2 n log(t ), and E t = {θ : θ ˆθ t Σt β}. We showed that w.p. 1 1/T 2 one has θ E t for all t [T ].
111 i.i.d. linear bandit after DHK08, RT10, AYPS11 Let β = 2 n log(t ), and E t = {θ : θ ˆθ t Σt β}. We showed that w.p. 1 1/T 2 one has θ E t for all t [T ]. The appropriate generalization of UCB is to select: ( θ t, a t ) = argmin (θ,a) E t A θ a (this optimization is NP-hard in general, more on that next slide).
112 i.i.d. linear bandit after DHK08, RT10, AYPS11 Let β = 2 n log(t ), and E t = {θ : θ ˆθ t Σt β}. We showed that w.p. 1 1/T 2 one has θ E t for all t [T ]. The appropriate generalization of UCB is to select: ( θ t, a t ) = argmin (θ,a) E t A θ a (this optimization is NP-hard in general, more on that next slide). Then one has on the high-probability event: T θ (a t a ) t=1 T (θ θ t ) a t β t=1 T t=1 a t Σ 1 t β T t a t 2. Σ 1 t
113 i.i.d. linear bandit after DHK08, RT10, AYPS11 Let β = 2 n log(t ), and E t = {θ : θ ˆθ t Σt β}. We showed that w.p. 1 1/T 2 one has θ E t for all t [T ]. The appropriate generalization of UCB is to select: ( θ t, a t ) = argmin (θ,a) E t A θ a (this optimization is NP-hard in general, more on that next slide). Then one has on the high-probability event: T θ (a t a ) t=1 T (θ θ t ) a t β t=1 T t=1 a t Σ 1 t To control the sum of squares we observe that: β T t a t 2. Σ 1 t det(σ t+1 ) = det(σ t ) det(i n +Σ 1/2 t a t (Σ 1/2 t a t ) ) = det(σ t )(1+ a t 2 ) Σ 1 t so that (assuming λ 1) log det(σ T +1 ) log det(σ 1 ) = t log(1+ a t 2 ) 1 Σ 1 t 2 t a t 2. Σ 1 t
114 i.i.d. linear bandit after DHK08, RT10, AYPS11 Let β = 2 n log(t ), and E t = {θ : θ ˆθ t Σt β}. We showed that w.p. 1 1/T 2 one has θ E t for all t [T ]. The appropriate generalization of UCB is to select: ( θ t, a t ) = argmin (θ,a) E t A θ a (this optimization is NP-hard in general, more on that next slide). Then one has on the high-probability event: T θ (a t a ) t=1 T (θ θ t ) a t β t=1 T t=1 a t Σ 1 t To control the sum of squares we observe that: β T t a t 2. Σ 1 t det(σ t+1 ) = det(σ t ) det(i n +Σ 1/2 t a t (Σ 1/2 t a t ) ) = det(σ t )(1+ a t 2 ) Σ 1 t so that (assuming λ 1) log det(σ T +1 ) log det(σ 1 ) = log(1+ a t 2 ) 1 a Σ 1 t 2. t 2 Σ 1 t t t Putting things together we see that the regret is O(n log(t ) T ).
115 What s the point of i.i.d. linear bandit? So far we did not get any real benefit from the i.i.d. assumption (the regret guarantee we obtained is the same as for the adversarial model). To me the key benefit is in the simplicity of the i.i.d. algorithm which makes it easy to incorporate further assumptions.
116 What s the point of i.i.d. linear bandit? So far we did not get any real benefit from the i.i.d. assumption (the regret guarantee we obtained is the same as for the adversarial model). To me the key benefit is in the simplicity of the i.i.d. algorithm which makes it easy to incorporate further assumptions. 1. Sparsity of θ: instead of regularization with l 2 -norm to define ˆθ one could regularize with l 1 -norm, see e.g., Johnson, Sivakumar and Banerjee [2016].
Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems, Part I. Sébastien Bubeck Theory Group
Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems, Part I Sébastien Bubeck Theory Group i.i.d. multi-armed bandit, Robbins [1952] i.i.d. multi-armed bandit, Robbins [1952] Known
More informationMulti-armed bandit models: a tutorial
Multi-armed bandit models: a tutorial CERMICS seminar, March 30th, 2016 Multi-Armed Bandit model: general setting K arms: for a {1,..., K}, (X a,t ) t N is a stochastic process. (unknown distributions)
More informationTwo generic principles in modern bandits: the optimistic principle and Thompson sampling
Two generic principles in modern bandits: the optimistic principle and Thompson sampling Rémi Munos INRIA Lille, France CSML Lunch Seminars, September 12, 2014 Outline Two principles: The optimistic principle
More informationAdvanced Machine Learning
Advanced Machine Learning Bandit Problems MEHRYAR MOHRI MOHRI@ COURANT INSTITUTE & GOOGLE RESEARCH. Multi-Armed Bandit Problem Problem: which arm of a K-slot machine should a gambler pull to maximize his
More informationBandit models: a tutorial
Gdt COS, December 3rd, 2015 Multi-Armed Bandit model: general setting K arms: for a {1,..., K}, (X a,t ) t N is a stochastic process. (unknown distributions) Bandit game: a each round t, an agent chooses
More informationOn the Complexity of Best Arm Identification in Multi-Armed Bandit Models
On the Complexity of Best Arm Identification in Multi-Armed Bandit Models Aurélien Garivier Institut de Mathématiques de Toulouse Information Theory, Learning and Big Data Simons Institute, Berkeley, March
More informationEASINESS IN BANDITS. Gergely Neu. Pompeu Fabra University
EASINESS IN BANDITS Gergely Neu Pompeu Fabra University EASINESS IN BANDITS Gergely Neu Pompeu Fabra University THE BANDIT PROBLEM Play for T rounds attempting to maximize rewards THE BANDIT PROBLEM Play
More informationOnline Learning and Online Convex Optimization
Online Learning and Online Convex Optimization Nicolò Cesa-Bianchi Università degli Studi di Milano N. Cesa-Bianchi (UNIMI) Online Learning 1 / 49 Summary 1 My beautiful regret 2 A supposedly fun game
More informationThe Online Approach to Machine Learning
The Online Approach to Machine Learning Nicolò Cesa-Bianchi Università degli Studi di Milano N. Cesa-Bianchi (UNIMI) Online Approach to ML 1 / 53 Summary 1 My beautiful regret 2 A supposedly fun game I
More informationOnline Learning and Sequential Decision Making
Online Learning and Sequential Decision Making Emilie Kaufmann CNRS & CRIStAL, Inria SequeL, emilie.kaufmann@univ-lille.fr Research School, ENS Lyon, Novembre 12-13th 2018 Emilie Kaufmann Sequential Decision
More informationOn Bayesian bandit algorithms
On Bayesian bandit algorithms Emilie Kaufmann joint work with Olivier Cappé, Aurélien Garivier, Nathaniel Korda and Rémi Munos July 1st, 2012 Emilie Kaufmann (Telecom ParisTech) On Bayesian bandit algorithms
More informationBandits for Online Optimization
Bandits for Online Optimization Nicolò Cesa-Bianchi Università degli Studi di Milano N. Cesa-Bianchi (UNIMI) Bandits for Online Optimization 1 / 16 The multiarmed bandit problem... K slot machines Each
More informationExponential Weights on the Hypercube in Polynomial Time
European Workshop on Reinforcement Learning 14 (2018) October 2018, Lille, France. Exponential Weights on the Hypercube in Polynomial Time College of Information and Computer Sciences University of Massachusetts
More informationReinforcement Learning
Reinforcement Learning Lecture 5: Bandit optimisation Alexandre Proutiere, Sadegh Talebi, Jungseul Ok KTH, The Royal Institute of Technology Objectives of this lecture Introduce bandit optimisation: the
More informationThe information complexity of sequential resource allocation
The information complexity of sequential resource allocation Emilie Kaufmann, joint work with Olivier Cappé, Aurélien Garivier and Shivaram Kalyanakrishan SMILE Seminar, ENS, June 8th, 205 Sequential allocation
More informationStratégies bayésiennes et fréquentistes dans un modèle de bandit
Stratégies bayésiennes et fréquentistes dans un modèle de bandit thèse effectuée à Telecom ParisTech, co-dirigée par Olivier Cappé, Aurélien Garivier et Rémi Munos Journées MAS, Grenoble, 30 août 2016
More informationExplore no more: Improved high-probability regret bounds for non-stochastic bandits
Explore no more: Improved high-probability regret bounds for non-stochastic bandits Gergely Neu SequeL team INRIA Lille Nord Europe gergely.neu@gmail.com Abstract This work addresses the problem of regret
More informationBandit Algorithms. Tor Lattimore & Csaba Szepesvári
Bandit Algorithms Tor Lattimore & Csaba Szepesvári Bandits Time 1 2 3 4 5 6 7 8 9 10 11 12 Left arm $1 $0 $1 $1 $0 Right arm $1 $0 Five rounds to go. Which arm would you play next? Overview What are bandits,
More informationYevgeny Seldin. University of Copenhagen
Yevgeny Seldin University of Copenhagen Classical (Batch) Machine Learning Collect Data Data Assumption The samples are independent identically distributed (i.i.d.) Machine Learning Prediction rule New
More informationAn Information-Theoretic Analysis of Thompson Sampling
Journal of Machine Learning Research (2015) Submitted ; Published An Information-Theoretic Analysis of Thompson Sampling Daniel Russo Department of Management Science and Engineering Stanford University
More informationStat 260/CS Learning in Sequential Decision Problems. Peter Bartlett
Stat 260/CS 294-102. Learning in Sequential Decision Problems. Peter Bartlett 1. Multi-armed bandit algorithms. Concentration inequalities. P(X ǫ) exp( ψ (ǫ))). Cumulant generating function bounds. Hoeffding
More informationOnline Learning with Feedback Graphs
Online Learning with Feedback Graphs Claudio Gentile INRIA and Google NY clagentile@gmailcom NYC March 6th, 2018 1 Content of this lecture Regret analysis of sequential prediction problems lying between
More informationSubsampling, Concentration and Multi-armed bandits
Subsampling, Concentration and Multi-armed bandits Odalric-Ambrym Maillard, R. Bardenet, S. Mannor, A. Baransi, N. Galichet, J. Pineau, A. Durand Toulouse, November 09, 2015 O-A. Maillard Subsampling and
More informationThe information complexity of best-arm identification
The information complexity of best-arm identification Emilie Kaufmann, joint work with Olivier Cappé and Aurélien Garivier MAB workshop, Lancaster, January th, 206 Context: the multi-armed bandit model
More informationComplex Bandit Problems and Thompson Sampling
Complex Bandit Problems and Aditya Gopalan Department of Electrical Engineering Technion, Israel aditya@ee.technion.ac.il Shie Mannor Department of Electrical Engineering Technion, Israel shie@ee.technion.ac.il
More informationInformational Confidence Bounds for Self-Normalized Averages and Applications
Informational Confidence Bounds for Self-Normalized Averages and Applications Aurélien Garivier Institut de Mathématiques de Toulouse - Université Paul Sabatier Thursday, September 12th 2013 Context Tree
More informationBayesian and Frequentist Methods in Bandit Models
Bayesian and Frequentist Methods in Bandit Models Emilie Kaufmann, Telecom ParisTech Bayes In Paris, ENSAE, October 24th, 2013 Emilie Kaufmann (Telecom ParisTech) Bayesian and Frequentist Bandits BIP,
More informationLecture 19: UCB Algorithm and Adversarial Bandit Problem. Announcements Review on stochastic multi-armed bandit problem
Lecture 9: UCB Algorithm and Adversarial Bandit Problem EECS598: Prediction and Learning: It s Only a Game Fall 03 Lecture 9: UCB Algorithm and Adversarial Bandit Problem Prof. Jacob Abernethy Scribe:
More informationExplore no more: Improved high-probability regret bounds for non-stochastic bandits
Explore no more: Improved high-probability regret bounds for non-stochastic bandits Gergely Neu SequeL team INRIA Lille Nord Europe gergely.neu@gmail.com Abstract This work addresses the problem of regret
More informationImproved Algorithms for Linear Stochastic Bandits
Improved Algorithms for Linear Stochastic Bandits Yasin Abbasi-Yadkori abbasiya@ualberta.ca Dept. of Computing Science University of Alberta Dávid Pál dpal@google.com Dept. of Computing Science University
More informationMultiple Identifications in Multi-Armed Bandits
Multiple Identifications in Multi-Armed Bandits arxiv:05.38v [cs.lg] 4 May 0 Sébastien Bubeck Department of Operations Research and Financial Engineering, Princeton University sbubeck@princeton.edu Tengyao
More informationBandits : optimality in exponential families
Bandits : optimality in exponential families Odalric-Ambrym Maillard IHES, January 2016 Odalric-Ambrym Maillard Bandits 1 / 40 Introduction 1 Stochastic multi-armed bandits 2 Boundary crossing probabilities
More informationTHE first formalization of the multi-armed bandit problem
EDIC RESEARCH PROPOSAL 1 Multi-armed Bandits in a Network Farnood Salehi I&C, EPFL Abstract The multi-armed bandit problem is a sequential decision problem in which we have several options (arms). We can
More informationSparse Linear Contextual Bandits via Relevance Vector Machines
Sparse Linear Contextual Bandits via Relevance Vector Machines Davis Gilton and Rebecca Willett Electrical and Computer Engineering University of Wisconsin-Madison Madison, WI 53706 Email: gilton@wisc.edu,
More informationCsaba Szepesvári 1. University of Alberta. Machine Learning Summer School, Ile de Re, France, 2008
LEARNING THEORY OF OPTIMAL DECISION MAKING PART I: ON-LINE LEARNING IN STOCHASTIC ENVIRONMENTS Csaba Szepesvári 1 1 Department of Computing Science University of Alberta Machine Learning Summer School,
More informationTwo optimization problems in a stochastic bandit model
Two optimization problems in a stochastic bandit model Emilie Kaufmann joint work with Olivier Cappé, Aurélien Garivier and Shivaram Kalyanakrishnan Journées MAS 204, Toulouse Outline From stochastic optimization
More informationarxiv: v4 [cs.lg] 22 Jul 2014
Learning to Optimize Via Information-Directed Sampling Daniel Russo and Benjamin Van Roy July 23, 2014 arxiv:1403.5556v4 cs.lg] 22 Jul 2014 Abstract We propose information-directed sampling a new algorithm
More informationOn the Complexity of Best Arm Identification with Fixed Confidence
On the Complexity of Best Arm Identification with Fixed Confidence Discrete Optimization with Noise Aurélien Garivier, Emilie Kaufmann COLT, June 23 th 2016, New York Institut de Mathématiques de Toulouse
More informationLearning to Optimize via Information-Directed Sampling
Learning to Optimize via Information-Directed Sampling Daniel Russo Stanford University Stanford, CA 94305 djrusso@stanford.edu Benjamin Van Roy Stanford University Stanford, CA 94305 bvr@stanford.edu
More informationBandit Convex Optimization: T Regret in One Dimension
Bandit Convex Optimization: T Regret in One Dimension arxiv:1502.06398v1 [cs.lg 23 Feb 2015 Sébastien Bubeck Microsoft Research sebubeck@microsoft.com Tomer Koren Technion tomerk@technion.ac.il February
More informationAnytime optimal algorithms in stochastic multi-armed bandits
Rémy Degenne LPMA, Université Paris Diderot Vianney Perchet CREST, ENSAE REMYDEGENNE@MATHUNIV-PARIS-DIDEROTFR VIANNEYPERCHET@NORMALESUPORG Abstract We introduce an anytime algorithm for stochastic multi-armed
More informationMinimax Policies for Combinatorial Prediction Games
Minimax Policies for Combinatorial Prediction Games Jean-Yves Audibert Imagine, Univ. Paris Est, and Sierra, CNRS/ENS/INRIA, Paris, France audibert@imagine.enpc.fr Sébastien Bubeck Centre de Recerca Matemàtica
More informationTutorial: PART 2. Online Convex Optimization, A Game- Theoretic Approach to Learning
Tutorial: PART 2 Online Convex Optimization, A Game- Theoretic Approach to Learning Elad Hazan Princeton University Satyen Kale Yahoo Research Exploiting curvature: logarithmic regret Logarithmic regret
More informationIntroduction to Bandit Algorithms. Introduction to Bandit Algorithms
Stochastic K-Arm Bandit Problem Formulation Consider K arms (actions) each correspond to an unknown distribution {ν k } K k=1 with values bounded in [0, 1]. At each time t, the agent pulls an arm I t {1,...,
More informationRegret in Online Combinatorial Optimization
Regret in Online Combinatorial Optimization Jean-Yves Audibert Imagine, Université Paris Est, and Sierra, CNRS/ENS/INRIA audibert@imagine.enpc.fr Sébastien Bubeck Department of Operations Research and
More informationBandit Algorithms. Zhifeng Wang ... Department of Statistics Florida State University
Bandit Algorithms Zhifeng Wang Department of Statistics Florida State University Outline Multi-Armed Bandits (MAB) Exploration-First Epsilon-Greedy Softmax UCB Thompson Sampling Adversarial Bandits Exp3
More informationNew Algorithms for Contextual Bandits
New Algorithms for Contextual Bandits Lev Reyzin Georgia Institute of Technology Work done at Yahoo! 1 S A. Beygelzimer, J. Langford, L. Li, L. Reyzin, R.E. Schapire Contextual Bandit Algorithms with Supervised
More informationStat 260/CS Learning in Sequential Decision Problems. Peter Bartlett
Stat 260/CS 294-102. Learning in Sequential Decision Problems. Peter Bartlett 1. Adversarial bandits Definition: sequential game. Lower bounds on regret from the stochastic case. Exp3: exponential weights
More informationBandit Optimization: Theory and Applications
1 / 40 Bandit Optimization: Theory and Applications Richard Combes 1 and Alexandre Proutière 2 1 Centrale-Supélec / L2S, France 2 KTH, Royal Institute of Technology, Sweden. SIGMETRICS 2015 2 / 40 Outline
More informationOrdinal Optimization and Multi Armed Bandit Techniques
Ordinal Optimization and Multi Armed Bandit Techniques Sandeep Juneja. with Peter Glynn September 10, 2014 The ordinal optimization problem Determining the best of d alternative designs for a system, on
More informationThe multi armed-bandit problem
The multi armed-bandit problem (with covariates if we have time) Vianney Perchet & Philippe Rigollet LPMA Université Paris Diderot ORFE Princeton University Algorithms and Dynamics for Games and Optimization
More informationIntroduction to Reinforcement Learning Part 3: Exploration for decision making, Application to games, optimization, and planning
Introduction to Reinforcement Learning Part 3: Exploration for decision making, Application to games, optimization, and planning Rémi Munos SequeL project: Sequential Learning http://researchers.lille.inria.fr/
More informationLearning to play K-armed bandit problems
Learning to play K-armed bandit problems Francis Maes 1, Louis Wehenkel 1 and Damien Ernst 1 1 University of Liège Dept. of Electrical Engineering and Computer Science Institut Montefiore, B28, B-4000,
More informationAn Estimation Based Allocation Rule with Super-linear Regret and Finite Lock-on Time for Time-dependent Multi-armed Bandit Processes
An Estimation Based Allocation Rule with Super-linear Regret and Finite Lock-on Time for Time-dependent Multi-armed Bandit Processes Prokopis C. Prokopiou, Peter E. Caines, and Aditya Mahajan McGill University
More informationOrdinal optimization - Empirical large deviations rate estimators, and multi-armed bandit methods
Ordinal optimization - Empirical large deviations rate estimators, and multi-armed bandit methods Sandeep Juneja Tata Institute of Fundamental Research Mumbai, India joint work with Peter Glynn Applied
More informationMulti-Armed Bandit Formulations for Identification and Control
Multi-Armed Bandit Formulations for Identification and Control Cristian R. Rojas Joint work with Matías I. Müller and Alexandre Proutiere KTH Royal Institute of Technology, Sweden ERNSI, September 24-27,
More informationarxiv: v2 [stat.ml] 19 Jul 2012
Thompson Sampling: An Asymptotically Optimal Finite Time Analysis Emilie Kaufmann, Nathaniel Korda and Rémi Munos arxiv:105.417v [stat.ml] 19 Jul 01 Telecom Paristech UMR CNRS 5141 & INRIA Lille - Nord
More informationFrom Bandits to Experts: A Tale of Domination and Independence
From Bandits to Experts: A Tale of Domination and Independence Nicolò Cesa-Bianchi Università degli Studi di Milano N. Cesa-Bianchi (UNIMI) Domination and Independence 1 / 1 From Bandits to Experts: A
More informationCOS 402 Machine Learning and Artificial Intelligence Fall Lecture 22. Exploration & Exploitation in Reinforcement Learning: MAB, UCB, Exp3
COS 402 Machine Learning and Artificial Intelligence Fall 2016 Lecture 22 Exploration & Exploitation in Reinforcement Learning: MAB, UCB, Exp3 How to balance exploration and exploitation in reinforcement
More informationOn the Complexity of Best Arm Identification with Fixed Confidence
On the Complexity of Best Arm Identification with Fixed Confidence Discrete Optimization with Noise Aurélien Garivier, joint work with Emilie Kaufmann CNRS, CRIStAL) to be presented at COLT 16, New York
More informationFull-information Online Learning
Introduction Expert Advice OCO LM A DA NANJING UNIVERSITY Full-information Lijun Zhang Nanjing University, China June 2, 2017 Outline Introduction Expert Advice OCO 1 Introduction Definitions Regret 2
More informationBandit Algorithms for Pure Exploration: Best Arm Identification and Game Tree Search. Wouter M. Koolen
Bandit Algorithms for Pure Exploration: Best Arm Identification and Game Tree Search Wouter M. Koolen Machine Learning and Statistics for Structures Friday 23 rd February, 2018 Outline 1 Intro 2 Model
More informationStochastic bandits: Explore-First and UCB
CSE599s, Spring 2014, Online Learning Lecture 15-2/19/2014 Stochastic bandits: Explore-First and UCB Lecturer: Brendan McMahan or Ofer Dekel Scribe: Javad Hosseini In this lecture, we like to answer this
More informationarxiv: v2 [cs.lg] 3 Nov 2012
arxiv:1204.5721v2 [cs.lg] 3 Nov 2012 Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems Sébastien Bubeck 1 and Nicolò Cesa-Bianchi 2 1 Department of Operations Research and Financial
More informationBandits and Exploration: How do we (optimally) gather information? Sham M. Kakade
Bandits and Exploration: How do we (optimally) gather information? Sham M. Kakade Machine Learning for Big Data CSE547/STAT548 University of Washington S. M. Kakade (UW) Optimization for Big data 1 / 22
More informationBandit Convex Optimization
March 7, 2017 Table of Contents 1 (BCO) 2 Projection Methods 3 Barrier Methods 4 Variance reduction 5 Other methods 6 Conclusion Learning scenario Compact convex action set K R d. For t = 1 to T : Predict
More informationRegional Multi-Armed Bandits
School of Information Science and Technology University of Science and Technology of China {wzy43, zrd7}@mail.ustc.edu.cn, congshen@ustc.edu.cn Abstract We consider a variant of the classic multiarmed
More informationBetter Algorithms for Benign Bandits
Better Algorithms for Benign Bandits Elad Hazan IBM Almaden 650 Harry Rd, San Jose, CA 95120 ehazan@cs.princeton.edu Satyen Kale Microsoft Research One Microsoft Way, Redmond, WA 98052 satyen.kale@microsoft.com
More informationLearning, Games, and Networks
Learning, Games, and Networks Abhishek Sinha Laboratory for Information and Decision Systems MIT ML Talk Series @CNRG December 12, 2016 1 / 44 Outline 1 Prediction With Experts Advice 2 Application to
More informationOnline Learning with Gaussian Payoffs and Side Observations
Online Learning with Gaussian Payoffs and Side Observations Yifan Wu 1 András György 2 Csaba Szepesvári 1 1 Department of Computing Science University of Alberta 2 Department of Electrical and Electronic
More informationThe Multi-Armed Bandit Problem
The Multi-Armed Bandit Problem Electrical and Computer Engineering December 7, 2013 Outline 1 2 Mathematical 3 Algorithm Upper Confidence Bound Algorithm A/B Testing Exploration vs. Exploitation Scientist
More informationStochastic Regret Minimization via Thompson Sampling
JMLR: Workshop and Conference Proceedings vol 35:1 22, 2014 Stochastic Regret Minimization via Thompson Sampling Sudipto Guha Department of Computer and Information Sciences, University of Pennsylvania,
More informationEfficient learning by implicit exploration in bandit problems with side observations
Efficient learning by implicit exploration in bandit problems with side observations Tomáš Kocák, Gergely Neu, Michal Valko, Rémi Munos SequeL team, INRIA Lille - Nord Europe, France SequeL INRIA Lille
More informationarxiv: v3 [cs.lg] 30 Jun 2012
arxiv:05874v3 [cslg] 30 Jun 0 Orly Avner Shie Mannor Department of Electrical Engineering, Technion Ohad Shamir Microsoft Research New England Abstract We consider a multi-armed bandit problem where the
More informationThe No-Regret Framework for Online Learning
The No-Regret Framework for Online Learning A Tutorial Introduction Nahum Shimkin Technion Israel Institute of Technology Haifa, Israel Stochastic Processes in Engineering IIT Mumbai, March 2013 N. Shimkin,
More informationLecture 4: Lower Bounds (ending); Thompson Sampling
CMSC 858G: Bandits, Experts and Games 09/12/16 Lecture 4: Lower Bounds (ending); Thompson Sampling Instructor: Alex Slivkins Scribed by: Guowei Sun,Cheng Jie 1 Lower bounds on regret (ending) Recap from
More informationKULLBACK-LEIBLER UPPER CONFIDENCE BOUNDS FOR OPTIMAL SEQUENTIAL ALLOCATION
Submitted to the Annals of Statistics arxiv: math.pr/0000000 KULLBACK-LEIBLER UPPER CONFIDENCE BOUNDS FOR OPTIMAL SEQUENTIAL ALLOCATION By Olivier Cappé 1, Aurélien Garivier 2, Odalric-Ambrym Maillard
More informationAn adaptive algorithm for finite stochastic partial monitoring
An adaptive algorithm for finite stochastic partial monitoring Gábor Bartók Navid Zolghadr Csaba Szepesvári Department of Computing Science, University of Alberta, AB, Canada, T6G E8 bartok@ualberta.ca
More informationA minimax and asymptotically optimal algorithm for stochastic bandits
Proceedings of Machine Learning Research 76:1 15, 017 Algorithmic Learning heory 017 A minimax and asymptotically optimal algorithm for stochastic bandits Pierre Ménard Aurélien Garivier Institut de Mathématiques
More informationThe Multi-Arm Bandit Framework
The Multi-Arm Bandit Framework A. LAZARIC (SequeL Team @INRIA-Lille) ENS Cachan - Master 2 MVA SequeL INRIA Lille MVA-RL Course In This Lecture A. LAZARIC Reinforcement Learning Algorithms Oct 29th, 2013-2/94
More informationMulti-Armed Bandit: Learning in Dynamic Systems with Unknown Models
c Qing Zhao, UC Davis. Talk at Xidian Univ., September, 2011. 1 Multi-Armed Bandit: Learning in Dynamic Systems with Unknown Models Qing Zhao Department of Electrical and Computer Engineering University
More informationFrom Bandits to Monte-Carlo Tree Search: The Optimistic Principle Applied to Optimization and Planning
From Bandits to Monte-Carlo Tree Search: The Optimistic Principle Applied to Optimization and Planning Rémi Munos To cite this version: Rémi Munos. From Bandits to Monte-Carlo Tree Search: The Optimistic
More informationCombinatorial Multi-Armed Bandit: General Framework, Results and Applications
Combinatorial Multi-Armed Bandit: General Framework, Results and Applications Wei Chen Microsoft Research Asia, Beijing, China Yajun Wang Microsoft Research Asia, Beijing, China Yang Yuan Computer Science
More informationImportance Weighting Without Importance Weights: An Efficient Algorithm for Combinatorial Semi-Bandits
Journal of Machine Learning Research 17 (2016 1-21 Submitted 3/15; Revised 7/16; Published 8/16 Importance Weighting Without Importance Weights: An Efficient Algorithm for Combinatorial Semi-Bandits Gergely
More informationThompson Sampling for Contextual Bandits with Linear Payoffs
Thompson Sampling for Contextual Bandits with Linear Payoffs Shipra Agrawal Microsoft Research Navin Goyal Microsoft Research Abstract Thompson Sampling is one of the oldest heuristics for multi-armed
More informationLecture 5: Regret Bounds for Thompson Sampling
CMSC 858G: Bandits, Experts and Games 09/2/6 Lecture 5: Regret Bounds for Thompson Sampling Instructor: Alex Slivkins Scribed by: Yancy Liao Regret Bounds for Thompson Sampling For each round t, we defined
More informationThe Multi-Armed Bandit Problem
Università degli Studi di Milano The bandit problem [Robbins, 1952]... K slot machines Rewards X i,1, X i,2,... of machine i are i.i.d. [0, 1]-valued random variables An allocation policy prescribes which
More informationOnline learning with feedback graphs and switching costs
Online learning with feedback graphs and switching costs A Proof of Theorem Proof. Without loss of generality let the independent sequence set I(G :T ) formed of actions (or arms ) from to. Given the sequence
More informationEfficient learning by implicit exploration in bandit problems with side observations
Efficient learning by implicit exploration in bandit problems with side observations omáš Kocák Gergely Neu Michal Valko Rémi Munos SequeL team, INRIA Lille Nord Europe, France {tomas.kocak,gergely.neu,michal.valko,remi.munos}@inria.fr
More informationRacing Thompson: an Efficient Algorithm for Thompson Sampling with Non-conjugate Priors
Racing Thompson: an Efficient Algorithm for Thompson Sampling with Non-conugate Priors Yichi Zhou 1 Jun Zhu 1 Jingwe Zhuo 1 Abstract Thompson sampling has impressive empirical performance for many multi-armed
More informationUniversity of Alberta. The Role of Information in Online Learning
University of Alberta The Role of Information in Online Learning by Gábor Bartók A thesis submitted to the Faculty of Graduate Studies and Research in partial fulfillment of the requirements for the degree
More informationMore Adaptive Algorithms for Adversarial Bandits
Proceedings of Machine Learning Research vol 75: 29, 208 3st Annual Conference on Learning Theory More Adaptive Algorithms for Adversarial Bandits Chen-Yu Wei University of Southern California Haipeng
More informationRevisiting the Exploration-Exploitation Tradeoff in Bandit Models
Revisiting the Exploration-Exploitation Tradeoff in Bandit Models joint work with Aurélien Garivier (IMT, Toulouse) and Tor Lattimore (University of Alberta) Workshop on Optimization and Decision-Making
More informationNew bounds on the price of bandit feedback for mistake-bounded online multiclass learning
Journal of Machine Learning Research 1 8, 2017 Algorithmic Learning Theory 2017 New bounds on the price of bandit feedback for mistake-bounded online multiclass learning Philip M. Long Google, 1600 Amphitheatre
More informationPure Exploration Stochastic Multi-armed Bandits
C&A Workshop 2016, Hangzhou Pure Exploration Stochastic Multi-armed Bandits Jian Li Institute for Interdisciplinary Information Sciences Tsinghua University Outline Introduction 2 Arms Best Arm Identification
More informationAnalysis of Thompson Sampling for the multi-armed bandit problem
Analysis of Thompson Sampling for the multi-armed bandit problem Shipra Agrawal Microsoft Research India shipra@microsoft.com avin Goyal Microsoft Research India navingo@microsoft.com Abstract We show
More informationPiecewise-stationary Bandit Problems with Side Observations
Jia Yuan Yu jia.yu@mcgill.ca Department Electrical and Computer Engineering, McGill University, Montréal, Québec, Canada. Shie Mannor shie.mannor@mcgill.ca; shie@ee.technion.ac.il Department Electrical
More informationTor Lattimore & Csaba Szepesvári
Bandit Algorithms Tor Lattimore & Csaba Szepesvári Outline 1 From Contextual to Linear Bandits 2 Stochastic Linear Bandits 3 Confidence Bounds for Least-Squares Estimators 4 Improved Regret for Fixed,
More informationExperts in a Markov Decision Process
University of Pennsylvania ScholarlyCommons Statistics Papers Wharton Faculty Research 2004 Experts in a Markov Decision Process Eyal Even-Dar Sham Kakade University of Pennsylvania Yishay Mansour Follow
More informationDynamic resource allocation: Bandit problems and extensions
Dynamic resource allocation: Bandit problems and extensions Aurélien Garivier Institut de Mathématiques de Toulouse MAD Seminar, Université Toulouse 1 October 3rd, 2014 The Bandit Model Roadmap 1 The Bandit
More information