An Overview of Recent and Brand New Primal-Dual Methods for Solving Convex Optimization Problems
|
|
- Norah Harmon
- 6 years ago
- Views:
Transcription
1 PGMO 1/32 An Overview of Recent and Brand New Primal-Dual Methods for Solving Convex Optimization Problems Emilie Chouzenoux Laboratoire d Informatique Gaspard Monge - CNRS Univ. Paris-Est Marne-la-Vallée, France 28 Oct. 2015
2 PGMO 2/32 In collaboration with A. Repetti J.-C. Pesquet
3 PGMO 3/32 Problem statement GOAL: Find a solution to the optimization problem minimize x H f(x) where H is a real Hilbertian space, f: H ],+ ] is convex. In the context of large scale problems, how to find an optimization algorithm able to deliver a reliable numerical solution in a reasonable time, with low memory requirement?
4 PGMO 4/32 Fundamental tools in convex analysis
5 PGMO 5/32 Notation and definitions Let f: H ],+ ]. The domain of function f is domf = {x H f(x) < + } If domf, function f is said to be proper. Function f is convex if ( (x,y) H 2 )( λ [0,1]) f(λx+(1 λ)y) λf(x)+(1 λ)f(y). Function f is lower semi-continuous (lsc) on H if, for every x H, for every sequence (x ) N of H, x x liminff(x ) f(x).
6 PGMO 5/32 Notation and definitions Let f: H ],+ ]. The domain of function f is domf = {x H f(x) < + } If domf, function f is said to be proper. Function f is convex if ( (x,y) H 2 )( λ [0,1]) f(λx+(1 λ)y) λf(x)+(1 λ)f(y). Function f is lower semi-continuous (lsc) on H if, for every x H, for every sequence (x ) N of H, x x liminff(x ) f(x). In the remainder of this tal, all functions will be assumed to be proper, lsc and convex.
7 PGMO 6/32 Notation and definitions The set of strongly positive, self adjoint, linear operators from H to H is denoted by S + (H). Let U S + (H). The weighted norm induced by U is with the convention = Id. U = U,
8 PGMO 6/32 Notation and definitions The set of strongly positive, self adjoint, linear operators from H to H is denoted by S + (H). Let U S + (H). The weighted norm induced by U is with the convention = Id. U = U, Let f: H ],+ [. Function f is said β-lipschitz differentiable if it is differentiable over H and its gradient fulfills with β ]0,+ [. ( (x,y) H 2 ) f(x) f(y) β x y,
9 Introduction Fundamental tools in convex analysis Primal-dual proximal algorithms Application to 3D mesh denoising PGMO 7/32 Subdifferential The subdifferential of f: H ],+ ] at x is the set f(x) = {t H ( y H) f(y) f(x)+ t y x } An element t of f(x) is called a subgradient of f at x. f(y) f(x)+ y x t t f(x) x x y If f is differentiable at x H then f(x) = { f(x)}.
10 Introduction Fundamental tools in convex analysis Primal-dual proximal algorithms Application to 3D mesh denoising PGMO 8/32 Conjugate function The conjugate of f: H ],+ ] is f : H [,+ ] such that ( u H) f ( ) (u) = sup x u f(x). x H f(x) x u f(x) x f (u) x f (u) x u ORIGINS: A. C. Clairaut ( ), A.-M. Legendre ( ), W. Fenchel ( )
11 PGMO 9/32 Notation and definitions The inf-convolution between f : H ],+ ] and g : H ],+ ] is f g: H [,+ ] x inf y H f(y)+g(x y). The convolution between f : H ],+ ] and g : H ],+ ] is f g: H [,+ ] x f(y)g(x y)dy. H PARTICULAR CASE: f ι {0} = f, where, for C H, ( x H) ι C (x) = { 0 if x C, + elsewhere.
12 PGMO 10/32 Proximity operator CHARACTERIZATION OF PROXIMITY OPERATOR ( x H) ŷ = prox U,f (x) x ŷ U 1 f(ŷ). The proximity operator prox U,f (x) of f at x H relative to the metric induced by U is the unique vector ŷ H such that f(ŷ)+ 1 2 ŷ x 2 U = inf f(y)+ 1 y x U(y x). y H 2
13 PGMO 11/32 Primal-dual proximal algorithms
14 PGMO 12/32 Primal-dual problem PRIMAL PROBLEM minimize x H f(x)+g(lx)+h(x). f: H ],+ ], g: G ],+ ], h: H R, β-lipschitz differentiable with β ]0,+ [, H and G real Hilbert spaces, L: H G linear and bounded.
15 PGMO 12/32 Primal-dual problem PRIMAL PROBLEM minimize x H f(x)+g(lx)+h(x). f: H ],+ ], g: G ],+ ], h: H R, β-lipschitz differentiable with β ]0,+ [, H and G real Hilbert spaces, L: H G linear and bounded. minimize v G (f h ) ( L v ) +g (v). DUAL PROBLEM
16 PGMO 13/32 Primal-dual problem KKT CONDITIONS If a pair ( x, v) fulfills the Karush-Kuhn-Tucer conditions: L v h( x) f( x) and L x g ( v), then { x is a solution to the primal problem v is a solution to the dual problem. LAGRANGIAN FORMULATION The problem is equivalent to searching for a saddle point of the Lagrangian: ( (x,y,v) H G 2 ) L(x,y,v) = f(x)+h(x)+g(y)+ v Lx y.
17 PGMO 14/32 Lin with Lagrange duality SADDLE POINTS ( x,ŷ, v) is a saddle point of the Lagrange function L if ( (x,y,v) H G 2 ) L( x,ŷ,v) L( x,ŷ, v) L(x,y, v). Let H and G be two real Hilbert spaces. Let ( x,ŷ, v) H G 2. Assume that ri(domg) L(domf) or domg ri ( L(domf) ). ( x,ŷ, v) is a saddle point of the Lagrange function ( x, v) is a Kuhn-Tucer point and ŷ = L x. EQUIVALENCE
18 PGMO 15/32 Primal-dual proximal algorithm for = 0,1,... y = prox τf (x τ ( h(x )+c +L ) ) v +a u = prox σg ( v +σl(2y x ) ) +b (x +1,v +1 ) = (x,v )+λ ((y,u ) (x,v ))
19 PGMO 15/32 Primal-dual proximal algorithm for = 0,1,... y = prox τf (x τ ( h(x )+c +L ) ) v +a u = prox σg ( v +σl(2y x ) ) +b (x +1,v +1 ) = (x,v )+λ ((y,u ) (x,v )) (τ,σ) ]0,+ [ 2 such that τ 1 σ L 2 > β/2,
20 PGMO 15/32 Primal-dual proximal algorithm for = 0,1,... y = prox τf (x τ ( h(x )+c +L ) ) v +a u = prox σg ( v +σl(2y x ) ) +b (x +1,v +1 ) = (x,v )+λ ((y,u ) (x,v )) (τ,σ) ]0,+ [ 2 such that τ 1 σ L 2 > β/2, for every N, λ ]0,λ[, where λ = 2 β(τ 1 σ L 2 ) 1 /2 [1,2[,
21 PGMO 15/32 Primal-dual proximal algorithm for = 0,1,... y = prox τf (x τ ( h(x )+c +L ) ) v +a u = prox σg ( v +σl(2y x ) ) +b (x +1,v +1 ) = (x,v )+λ ((y,u ) (x,v )) (τ,σ) ]0,+ [ 2 such that τ 1 σ L 2 > β/2, for every N, λ ]0,λ[, where λ = 2 β(τ 1 σ L 2 ) 1 /2 [1,2[, a < +, N N b < +, and N c < +,
22 PGMO 15/32 Primal-dual proximal algorithm for = 0,1,... y = prox τf (x τ ( h(x )+c +L ) ) v +a u = prox σg ( v +σl(2y x ) ) +b (x +1,v +1 ) = (x,v )+λ ((y,u ) (x,v )) (τ,σ) ]0,+ [ 2 such that τ 1 σ L 2 > β/2, for every N, λ ]0,λ[, where λ = 2 β(τ 1 σ L 2 ) 1 /2 [1,2[, a < +, N N b < +, and N c < +, there exists x H such that 0 f(x)+ h(x)+l g ( Lx ), Then: x x where x is a solution to the primal problem, v v where v is a solution to the dual problem.
23 PGMO 16/32 Preconditioned primal-dual proximal algorithm for = 0,1,... ( y = prox W 1,f x W ( h(x )+c +L ) ) v +a ( u = prox U 1,g v +UL(2y x ) ) +b (x +1,v +1 ) = (x,v )+λ ((y,u ) (x,v )) W S + (H) and U S + (G) such that 1 U 1/2 LW 1/2 > β 2 W, for every N, λ ]0,1], and inf N λ > 0, a < +, N N b < +, and N c < +, there exists x H such that 0 f(x)+ h(x)+l g ( Lx ). Then: x x where x is a solution to the primal problem, v v where v is a solution to the dual problem.
24 PGMO 17/32 Parallel primal-dual proximal algorithm PRIMAL PROBLEM minimize x H f(x)+g(lx)+h(x).
25 PGMO 17/32 Parallel primal-dual proximal algorithm PRIMAL PROBLEM minimize x H q f(x)+ g n (L n x)+h(x). n=1 with, for all n {1,...,q}, g n : G n ],+ ], (G n ) 1 n q real Hilbert spaces such thatg = G 1 G q, andl n : H G n bounded linear operators.
26 PGMO 17/32 Parallel primal-dual proximal algorithm PRIMAL PROBLEM minimize x H q f(x)+ g n (L n x)+h(x). n=1 with, for all n {1,...,q}, g n : G n ],+ ], (G n ) 1 n q real Hilbert spaces such thatg = G 1 G q, andl n : H G n bounded linear operators. minimize (v (1),...,v (q) ) G (f h ) ( q L nv (n)) + n=1 DUAL PROBLEM q gn(v (n) ). n=1
27 PGMO 17/32 Parallel primal-dual proximal algorithm PRIMAL PROBLEM minimize x H q f(x)+ g n (L n x)+h(x). n=1 with, for all n {1,...,q}, g n : G n ],+ ], (G n ) 1 n q real Hilbert spaces such thatg = G 1 G q, andl n : H G n bounded linear operators. minimize (v (1),...,v (q) ) G (f h ) ( q L nv (n)) + n=1 DUAL PROBLEM q gn(v (n) ). The separability of g implies that prox g can be obtained by computing in parallel the proximal operators of functions (g n) 1 n q PARALLEL PRIMAL-DUAL SCHEMES. n=1
28 PGMO 18/32 Parallel primal-dual proximal algorithm for = 0,1,... ( y = prox W 1,f x W ( q h(x )+c + n=1 x +1 = x +λ (y x ) for n,...,q u (n) ( (n) = prox U 1 n,g v n +U n L n (2y x ) ) +b (n) v (n) +1 = v(n) +λ (u (n) v (n) ) L nv (n) ) ) +a
29 PGMO 19/32 Bibliographical remars Pioneering wor: Arrow-Hurwicz-Uzawa method for saddle point problems [Arrow et al ] [Nedić and Ozdaglar ] Methods based on Forward-Bacward iteration: type I : [Vu ][Condat ] (extensions of [Esser et al ][Chambolle and Poc ]) type II: [Combettes et al ] (extensions of [Loris and Verhoeven ][Chen et al ]) Methods based on Forward-Bacward-Forward iteration [Combettes and Pesquet ] [Boţ and Hendrich,2014] Projection based methods... [Alotaibi et al ]
30 PGMO 20/32 Random bloc coordinate strategy
31 PGMO 21/32 Acceleration via bloc alternation Idea: variable splitting.
32 PGMO 21/32 Acceleration via bloc alternation Idea: variable splitting. x (1) H 1 x (2) H 2 g x H H = p j=1 H j x (p) H p H 1,...,H p are separable real Hilbert spaces
33 PGMO 21/32 Acceleration via bloc alternation Simplifying assumption: f and h are bloc separable functions. x (1) x (2) f x = f = p f j (x (j) ) j=1 x (p)
34 PGMO 21/32 Acceleration via bloc alternation Simplifying assumption: f and h are bloc separable functions. x (1) x (2) h x = h = p h j (x (j) ) j=1 x (p) ( j {1,...,p}) h j β j -Lipschitz differentiable with β j ]0,+ [.
35 PGMO 22/32 Acceleration via bloc alternation For each iteration N, update a subset of components ( Gauss-Seidel methods). ADVANTAGES: Reduced computational cost per iteration, Reduced memory requirement, More flexibility.
36 PGMO 23/32 Primal-dual problem PRIMAL PROBLEM Let F the set of solutions to the problem p ( ) minimize f j (x (j) )+h j (x (j) ) x (1) H 1,...,x (p) H p j=1 ( j {1,...,p})( n {1,...,q}) H j and G n separable real Hilbert spaces, f j: H j ],+ ] + q ( p g n L n,j x (j)) n=1 h j: H j R β j-lipschitz differentiable, with β j ]0,+ [ g n: G n ],+ ] L n,j: H j G n linear and bounded, L n = { j {1,...,p} L n,j 0 }, et L j = { n {1,...,q} L n,j 0 }. j=1
37 PGMO 23/32 Primal-dual problem PRIMAL PROBLEM Let F the set of solutions to the problem p ( ) minimize f j (x (j) )+h j (x (j) ) x (1) H 1,...,x (p) H p j=1 + q ( p g n L n,j x (j)) Let F DUAL PROBLEM the set of solutions to the problem p ( q minimize (fj h j) L n,jv (n)) q ( ) + gn(v (n) ) v (1) G 1,...,v (q) G q j=1 n=1 n=1 n=1 j=1 Assume that there exists (x (1),...,x (p) ) H 1... H p such that q ( j {1,...,p}) 0 f j (x (j) )+ h j (x (j) )+ L ( n,j g n Ln,j x (j)). n=1 GOAL: Find an F F -valued random variable ( x, v).
38 PGMO 24/32 Random bloc coordinate primal-dual proximal algorithm For = 0,1,... for j,...,p ( (j) y = ε (j) for n,...,q ( u(n) = ε (p+n) v (n) +1 = v(n) ( (j) prox W 1 j,f j x W j( h j (x (j) )+c(j) + q n=1 L n,j v(n) )) +a (j) x (j) +1 = x(j) +λ ε (j) (y(j) x (j) ) prox U 1 n,g n +λ ε (p+n) ( v (n) p +U n j=1 (u (n) v (n) ), ) L n,j (2y (j) x (j) )) +b (n) ) (ε ) N is a vector of p+q Boolean variables randomly chosen at each iteration N in order to signal the primal and dual blocs to be updated.
39 PGMO 25/32 Random bloc coordinate primal-dual proximal algorithm Assume that The variables (ε ) N are independent and identically distributed, such that for every N, ( n {1,...,q}) P[ε (p+n) ] > 0, and ( j {1,...,p}) ε (j) { (p+n) = max ε n L j}, 1 n q
40 PGMO 25/32 Random bloc coordinate primal-dual proximal algorithm Assume that The variables (ε ) N are independent and identically distributed, such that for every N, ( n {1,...,q}) P[ε (p+n) ] > 0, and ( j {1,...,p}) ε (j) { (p+n) = max ε n L j}, 1 n q ( j {1,...,p}) W j S + (H j), ( n {1,...,q}) U n S + (G n), with ( p ) 1/2 1 q j=1 n=1 U1/2 n L n,jw 1/2 j 2 > 1 max{( Wj βj) 2 1 j p}, ( N) λ ]0,1] and inf N λ > 0, E a 2 < +, E b 2 < +, and E c 2 < + a.s. N N N (x ) N converges wealy a.s. to an F-valued random variable. (v ) N converges wealy a.s. to an F -valued random variable. Proof: Based on properties of quasi-fejér stochastic sequences [Combettes & Pesquet 2015].
41 PGMO 26/32 Illustration of the random sampling strategy Bloc selection ( N) x (1) activated when ε (1) How to choose ( N) the variable ε = (ε (1),...,ε(6) )? x (2) activated when ε (2) x (3) activated when ε (3) x (4) activated when ε (4) x (5) activated when ε (5) x (6) activated when ε (6)
42 PGMO 26/32 Illustration of the random sampling strategy Bloc selection ( N) x (1) activated when ε (1) x (2) activated when ε (2) x (3) activated when ε (3) How to choose ( N) the variable ε = (ε (1),...,ε(6) )? P[ε = (1,1,0,0,0,0)] = 0.1 x (4) activated when ε (4) x (5) activated when ε (5) x (6) activated when ε (6)
43 PGMO 26/32 Illustration of the random sampling strategy Bloc selection ( n N) x (1) activated when ε (1) x (2) activated when ε (2) x (3) activated when ε (3) How to choose ( N) the variable ε = (ε (1),...,ε(6) )? P[ε = (1,1,0,0,0,0)] = 0.1 P[ε = (1,0,1,0,0,0)] = 0.2 x (4) activated when ε (4) x (5) activated when ε (5) x (6) activated when ε (6)
44 PGMO 26/32 Illustration of the random sampling strategy Bloc selection ( N) x (1) activated when ε (1) x (2) activated when ε (2) x (3) activated when ε (3) x (4) activated when ε (4) How to choose ( N) the variable ε = (ε (1),...,ε(6) )? P[ε = (1,1,0,0,0,0)] = 0.1 P[ε = (1,0,1,0,0,0)] = 0.2 P[ε = (1,0,0,1,1,0)] = 0.2 x (5) activated when ε (5) x (6) activated when ε (6)
45 PGMO 26/32 Illustration of the random sampling strategy Bloc selection ( N) x (1) activated when ε (1) x (2) activated when ε (2) x (3) activated when ε (3) x (4) activated when ε (4) x (5) activated when ε (5) How to choose ( N) the variable ε = (ε (1),...,ε(6) )? P[ε = (1,1,0,0,0,0)] = 0.1 P[ε = (1,0,1,0,0,0)] = 0.2 P[ε = (1,0,0,1,1,0)] = 0.2 P[ε = (0,1,1,1,1,1)] = 0.5 x (6) activated when ε (6)
46 PGMO 27/32 Extension to distributed algorithms... OPTIMIZATION PROBLEM Find an element of the set F of the solutions to m minimize f i (x)+h i (x)+g i (M i x) x H i=1 with H,G 1,...,G m separable Hilbert spaces and, for all i {1,...,m}: h i β i -Lipschitz differentiable with β i ]0,+ [, M i : H G i non-zero linear and bounded operators.
47 PGMO 27/32 Extension to distributed algorithms... OPTIMIZATION PROBLEM Find an element of the set F of the solutions to m minimize f i (x)+h i (x)+g i (M i x) x H i=1 OPTIMIZATION PROBLEM Find (x 1,...,x m ) H m such that m minimize f i (x i )+h i (x i )+g i (M i x i ) (x i ) 1 i m Λ m i=1 with Λ m = { (x i ) 1 i m H m } x 1 =... = x m
48 PGMO 28/32 3D mesh denoising (ANR GRAPHSIP) Original mesh x Observed mesh z Undirected nonreflexive graph OBJECTIVE: Estimate x = (x i ) 1 i M from noisy observations z = (z i ) 1 i M where, for every i {1,...,M}, x i R 3 is the vector of 3D coordinates of the i-th vertex of a mesh H = (R 3 ) M
49 PGMO 28/32 3D mesh denoising (ANR GRAPHSIP) OBJECTIVE: Estimate x = (x i ) 1 i M from noisy observations z = (z i ) 1 i M where, for every i {1,...,M}, x i R 3 is the vector of 3D coordinates of the i-th vertex of a mesh COST FUNCTION: Φ(x) = M ψ j (x j z j )+ι Cj (x j )+η j (x j x i ) i Nj 1,2 j=1 where ( j {1,...,M}), ψ j : R 3 R: l 2 l 1 Huber function robust data fidelity measure convex, Lipschitz differentiable function C j : nonempty convex subset of R 3 N j : neighborhood of j-th vertex (η j ) 1 j M : nonnegative regularization constants.
50 PGMO 28/32 3D mesh denoising (ANR GRAPHSIP) OBJECTIVE: Estimate x = (x i) 1 i M from noisy observations z = (z i) 1 i M where, for every i {1,...,M}, x i R 3 is the vector of 3D coordinates of the i-th vertex of a mesh COST FUNCTION: M Φ(x) = ψ j(x j z j)+ι Cj (x j)+η j (x j x i) i Nj 1,2 j=1 IMPLEMENTATION DETAILS: a bloc a vertex p = q = M ( j {1,...,M}) h j = ψ j( z j) f j = ι Cj ( {1,...,M})( x R 3M ) g (L x) = (x x i) i N 1,2
51 PGMO 29/32 Simulation results positions of the original mesh are corrupted through an additive i.i.d. zero-mean Gaussian mixture noise model. a limited number r of variables can be handled at each iteration, where p ε j,n = r p. j=1 mesh decomposed into p/r non-overlapping sets. Original mesh, M Noisy mesh, MSE =
52 PGMO 29/32 Simulation results Proposed reconstruction MSE = Laplacian smoothing MSE =
53 PGMO 30/32 Complexity Time (s.) p/r Memory (Mb) dashed line: required memory continuous line: reconstruction time
54 PGMO 31/32 Conclusion Simple handling of Proven convergence and robustness to errors Primal-dual proximal algorithms Resolution of (non) smooth convex optimization problems contraints No inversion of linear operators Proven efficiency in some signal /image processing applications Bloccoordinate strategies Flexibility in the random selection of primal/dual components Distributed versions of the algorithms available
55 PGMO 32/32 Some references... F. Abboud, E. Chouzenoux, J.-C. Pesquet, J.-H. Chenot and L. Laborelli A Dual Bloc Coordinate Proximal Algorithm with Application to Deconvolution of Interlaced Video Sequences IEEE International Conference on Image Processing (ICIP 2015), 5 p., Québec, Canada, Sep G. Chierchia, N. Pustelni, B. Pesquet-Popescu and J.-C. Pesquet A Non-Local Structure Tensor Based Approach for Multicomponent Image Recovery Problems IEEE Transaction on Image Processing, 23(12), pp , Dec P. L. Combettes and J.-C. Pesquet Proximal Splitting methods in Signal Processing in Fixed-Point Algorithms for Inverse Problems in Science and Engineering, H. H. Bausche, R. Burachi, P. L. Combettes, V. Elser, D. R. Lue, and H. Wolowicz editors. Springer-Verlag, New Yor, pp , P. L. Combettes, L. Condat, J.-C. Pesquet, and B. C. Vũ A Forward-Bacward View of Some Primal-Dual Optimization Methods in Image Recovery IEEE International Conference on Image Processing (ICIP 2014), 5 p., Paris, France, Oct , P. Combettes and J.-C Pesquet Stochastic Quasi-Fejér Bloc-Coordinate Fixed Point Iterations with Random Sweeping SIAM Journal on Optimization, 25(2), pp , A. Florescu, E. Chouzenoux, J.-C. Pesquet, P. Ciuciu and S. Ciochina A Majorize-Minimize Memory Gradient Method for Complex-Valued Inverse Problems Signal Processing, 103, pp , A. Jeziersa, E. Chouzenoux, J.-C. Pesquet and H. Talbot A Convex Approach for Image Restoration with Exact Poisson-Gaussian Lielihood to appear in SIAM Journal in Imaging Sciences, N. Komodais and J.-C. Pesquet Playing with Duality: An Overview of Recent Primal-Dual Approaches for Solving Large-Scale Optimization Problems to appear in IEEE Signal Processing Magazine, J.-C. Pesquet and A. Repetti A Class of Randomized Primal-Dual Algorithms for Distributed Optimization to appear in Journal of Nonlinear and Convex Analysis, A. Repetti, E. Chouzenoux and J.-C. Pesquet A Random Bloc-Coordinate Primal-Dual Proximal Algorithm with Application to 3D Mesh Denoising IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2015), 5 p., Brisbane, Australia, Apr , 2015.
A Parallel Block-Coordinate Approach for Primal-Dual Splitting with Arbitrary Random Block Selection
EUSIPCO 2015 1/19 A Parallel Block-Coordinate Approach for Primal-Dual Splitting with Arbitrary Random Block Selection Jean-Christophe Pesquet Laboratoire d Informatique Gaspard Monge - CNRS Univ. Paris-Est
More informationIn collaboration with J.-C. Pesquet A. Repetti EC (UPE) IFPEN 16 Dec / 29
A Random block-coordinate primal-dual proximal algorithm with application to 3D mesh denoising Emilie CHOUZENOUX Laboratoire d Informatique Gaspard Monge - CNRS Univ. Paris-Est, France Horizon Maths 2014
More informationSignal Processing and Networks Optimization Part VI: Duality
Signal Processing and Networks Optimization Part VI: Duality Pierre Borgnat 1, Jean-Christophe Pesquet 2, Nelly Pustelnik 1 1 ENS Lyon Laboratoire de Physique CNRS UMR 5672 pierre.borgnat@ens-lyon.fr,
More informationarxiv: v1 [math.oc] 20 Jun 2014
A forward-backward view of some primal-dual optimization methods in image recovery arxiv:1406.5439v1 [math.oc] 20 Jun 2014 P. L. Combettes, 1 L. Condat, 2 J.-C. Pesquet, 3 and B. C. Vũ 4 1 Sorbonne Universités
More informationVariable Metric Forward-Backward Algorithm
Variable Metric Forward-Backward Algorithm 1/37 Variable Metric Forward-Backward Algorithm for minimizing the sum of a differentiable function and a convex function E. Chouzenoux in collaboration with
More informationOn the convergence rate of a forward-backward type primal-dual splitting algorithm for convex optimization problems
On the convergence rate of a forward-backward type primal-dual splitting algorithm for convex optimization problems Radu Ioan Boţ Ernö Robert Csetnek August 5, 014 Abstract. In this paper we analyze the
More informationSplitting Techniques in the Face of Huge Problem Sizes: Block-Coordinate and Block-Iterative Approaches
Splitting Techniques in the Face of Huge Problem Sizes: Block-Coordinate and Block-Iterative Approaches Patrick L. Combettes joint work with J.-C. Pesquet) Laboratoire Jacques-Louis Lions Faculté de Mathématiques
More informationADMM for monotone operators: convergence analysis and rates
ADMM for monotone operators: convergence analysis and rates Radu Ioan Boţ Ernö Robert Csetne May 4, 07 Abstract. We propose in this paper a unifying scheme for several algorithms from the literature dedicated
More informationA Primal-dual Three-operator Splitting Scheme
Noname manuscript No. (will be inserted by the editor) A Primal-dual Three-operator Splitting Scheme Ming Yan Received: date / Accepted: date Abstract In this paper, we propose a new primal-dual algorithm
More informationPrimal-dual algorithms for the sum of two and three functions 1
Primal-dual algorithms for the sum of two and three functions 1 Ming Yan Michigan State University, CMSE/Mathematics 1 This works is partially supported by NSF. optimization problems for primal-dual algorithms
More informationOn the acceleration of the double smoothing technique for unconstrained convex optimization problems
On the acceleration of the double smoothing technique for unconstrained convex optimization problems Radu Ioan Boţ Christopher Hendrich October 10, 01 Abstract. In this article we investigate the possibilities
More informationA memory gradient algorithm for l 2 -l 0 regularization with applications to image restoration
A memory gradient algorithm for l 2 -l 0 regularization with applications to image restoration E. Chouzenoux, A. Jezierska, J.-C. Pesquet and H. Talbot Université Paris-Est Lab. d Informatique Gaspard
More informationConvergence analysis for a primal-dual monotone + skew splitting algorithm with applications to total variation minimization
Convergence analysis for a primal-dual monotone + skew splitting algorithm with applications to total variation minimization Radu Ioan Boţ Christopher Hendrich November 7, 202 Abstract. In this paper we
More informationA Majorize-Minimize subspace approach for l 2 -l 0 regularization with applications to image processing
A Majorize-Minimize subspace approach for l 2 -l 0 regularization with applications to image processing Emilie Chouzenoux emilie.chouzenoux@univ-mlv.fr Université Paris-Est Lab. d Informatique Gaspard
More informationarxiv: v4 [math.oc] 29 Jan 2018
Noname manuscript No. (will be inserted by the editor A new primal-dual algorithm for minimizing the sum of three functions with a linear operator Ming Yan arxiv:1611.09805v4 [math.oc] 29 Jan 2018 Received:
More informationBASICS OF CONVEX ANALYSIS
BASICS OF CONVEX ANALYSIS MARKUS GRASMAIR 1. Main Definitions We start with providing the central definitions of convex functions and convex sets. Definition 1. A function f : R n R + } is called convex,
More informationDual methods for the minimization of the total variation
1 / 30 Dual methods for the minimization of the total variation Rémy Abergel supervisor Lionel Moisan MAP5 - CNRS UMR 8145 Different Learning Seminar, LTCI Thursday 21st April 2016 2 / 30 Plan 1 Introduction
More informationAccelerated Dual Gradient-Based Methods for Total Variation Image Denoising/Deblurring Problems (and other Inverse Problems)
Accelerated Dual Gradient-Based Methods for Total Variation Image Denoising/Deblurring Problems (and other Inverse Problems) Donghwan Kim and Jeffrey A. Fessler EECS Department, University of Michigan
More informationConvex Optimization Theory. Chapter 5 Exercises and Solutions: Extended Version
Convex Optimization Theory Chapter 5 Exercises and Solutions: Extended Version Dimitri P. Bertsekas Massachusetts Institute of Technology Athena Scientific, Belmont, Massachusetts http://www.athenasc.com
More informationVictoria Martín-Márquez
A NEW APPROACH FOR THE CONVEX FEASIBILITY PROBLEM VIA MONOTROPIC PROGRAMMING Victoria Martín-Márquez Dep. of Mathematical Analysis University of Seville Spain XIII Encuentro Red de Análisis Funcional y
More informationAdaptive discretization and first-order methods for nonsmooth inverse problems for PDEs
Adaptive discretization and first-order methods for nonsmooth inverse problems for PDEs Christian Clason Faculty of Mathematics, Universität Duisburg-Essen joint work with Barbara Kaltenbacher, Tuomo Valkonen,
More informationMath 273a: Optimization Convex Conjugacy
Math 273a: Optimization Convex Conjugacy Instructor: Wotao Yin Department of Mathematics, UCLA Fall 2015 online discussions on piazza.com Convex conjugate (the Legendre transform) Let f be a closed proper
More informationarxiv: v1 [math.oc] 12 Mar 2013
On the convergence rate improvement of a primal-dual splitting algorithm for solving monotone inclusion problems arxiv:303.875v [math.oc] Mar 03 Radu Ioan Boţ Ernö Robert Csetnek André Heinrich February
More informationI P IANO : I NERTIAL P ROXIMAL A LGORITHM FOR N ON -C ONVEX O PTIMIZATION
I P IANO : I NERTIAL P ROXIMAL A LGORITHM FOR N ON -C ONVEX O PTIMIZATION Peter Ochs University of Freiburg Germany 17.01.2017 joint work with: Thomas Brox and Thomas Pock c 2017 Peter Ochs ipiano c 1
More informationBrøndsted-Rockafellar property of subdifferentials of prox-bounded functions. Marc Lassonde Université des Antilles et de la Guyane
Conference ADGO 2013 October 16, 2013 Brøndsted-Rockafellar property of subdifferentials of prox-bounded functions Marc Lassonde Université des Antilles et de la Guyane Playa Blanca, Tongoy, Chile SUBDIFFERENTIAL
More informationMonotone Operator Splitting Methods in Signal and Image Recovery
Monotone Operator Splitting Methods in Signal and Image Recovery P.L. Combettes 1, J.-C. Pesquet 2, and N. Pustelnik 3 2 Univ. Pierre et Marie Curie, Paris 6 LJLL CNRS UMR 7598 2 Univ. Paris-Est LIGM CNRS
More informationAbout Split Proximal Algorithms for the Q-Lasso
Thai Journal of Mathematics Volume 5 (207) Number : 7 http://thaijmath.in.cmu.ac.th ISSN 686-0209 About Split Proximal Algorithms for the Q-Lasso Abdellatif Moudafi Aix Marseille Université, CNRS-L.S.I.S
More informationConvex Optimization and Modeling
Convex Optimization and Modeling Duality Theory and Optimality Conditions 5th lecture, 12.05.2010 Jun.-Prof. Matthias Hein Program of today/next lecture Lagrangian and duality: the Lagrangian the dual
More informationADMM and Fast Gradient Methods for Distributed Optimization
ADMM and Fast Gradient Methods for Distributed Optimization João Xavier Instituto Sistemas e Robótica (ISR), Instituto Superior Técnico (IST) European Control Conference, ECC 13 July 16, 013 Joint work
More informationA Dykstra-like algorithm for two monotone operators
A Dykstra-like algorithm for two monotone operators Heinz H. Bauschke and Patrick L. Combettes Abstract Dykstra s algorithm employs the projectors onto two closed convex sets in a Hilbert space to construct
More informationProximal splitting methods on convex problems with a quadratic term: Relax!
Proximal splitting methods on convex problems with a quadratic term: Relax! The slides I presented with added comments Laurent Condat GIPSA-lab, Univ. Grenoble Alpes, France Workshop BASP Frontiers, Jan.
More informationSparse Optimization Lecture: Dual Methods, Part I
Sparse Optimization Lecture: Dual Methods, Part I Instructor: Wotao Yin July 2013 online discussions on piazza.com Those who complete this lecture will know dual (sub)gradient iteration augmented l 1 iteration
More informationA Tutorial on Primal-Dual Algorithm
A Tutorial on Primal-Dual Algorithm Shenlong Wang University of Toronto March 31, 2016 1 / 34 Energy minimization MAP Inference for MRFs Typical energies consist of a regularization term and a data term.
More informationSome Inexact Hybrid Proximal Augmented Lagrangian Algorithms
Some Inexact Hybrid Proximal Augmented Lagrangian Algorithms Carlos Humes Jr. a, Benar F. Svaiter b, Paulo J. S. Silva a, a Dept. of Computer Science, University of São Paulo, Brazil Email: {humes,rsilva}@ime.usp.br
More informationIterative Convex Optimization Algorithms; Part One: Using the Baillon Haddad Theorem
Iterative Convex Optimization Algorithms; Part One: Using the Baillon Haddad Theorem Charles Byrne (Charles Byrne@uml.edu) http://faculty.uml.edu/cbyrne/cbyrne.html Department of Mathematical Sciences
More informationPrimal-dual coordinate descent
Primal-dual coordinate descent Olivier Fercoq Joint work with P. Bianchi & W. Hachem 15 July 2015 1/28 Minimize the convex function f, g, h convex f is differentiable Problem min f (x) + g(x) + h(mx) x
More informationGauge optimization and duality
1 / 54 Gauge optimization and duality Junfeng Yang Department of Mathematics Nanjing University Joint with Shiqian Ma, CUHK September, 2015 2 / 54 Outline Introduction Duality Lagrange duality Fenchel
More informationPrimal-dual coordinate descent A Coordinate Descent Primal-Dual Algorithm with Large Step Size and Possibly Non-Separable Functions
Primal-dual coordinate descent A Coordinate Descent Primal-Dual Algorithm with Large Step Size and Possibly Non-Separable Functions Olivier Fercoq and Pascal Bianchi Problem Minimize the convex function
More informationDual Decomposition.
1/34 Dual Decomposition http://bicmr.pku.edu.cn/~wenzw/opt-2017-fall.html Acknowledgement: this slides is based on Prof. Lieven Vandenberghes lecture notes Outline 2/34 1 Conjugate function 2 introduction:
More informationON GAP FUNCTIONS OF VARIATIONAL INEQUALITY IN A BANACH SPACE. Sangho Kum and Gue Myung Lee. 1. Introduction
J. Korean Math. Soc. 38 (2001), No. 3, pp. 683 695 ON GAP FUNCTIONS OF VARIATIONAL INEQUALITY IN A BANACH SPACE Sangho Kum and Gue Myung Lee Abstract. In this paper we are concerned with theoretical properties
More informationSelf-dual Smooth Approximations of Convex Functions via the Proximal Average
Chapter Self-dual Smooth Approximations of Convex Functions via the Proximal Average Heinz H. Bauschke, Sarah M. Moffat, and Xianfu Wang Abstract The proximal average of two convex functions has proven
More informationAdaptive Primal Dual Optimization for Image Processing and Learning
Adaptive Primal Dual Optimization for Image Processing and Learning Tom Goldstein Rice University tag7@rice.edu Ernie Esser University of British Columbia eesser@eos.ubc.ca Richard Baraniuk Rice University
More informationMaximal monotone operators are selfdual vector fields and vice-versa
Maximal monotone operators are selfdual vector fields and vice-versa Nassif Ghoussoub Department of Mathematics, University of British Columbia, Vancouver BC Canada V6T 1Z2 nassif@math.ubc.ca February
More informationSolving monotone inclusions involving parallel sums of linearly composed maximally monotone operators
Solving monotone inclusions involving parallel sums of linearly composed maximally monotone operators Radu Ioan Boţ Christopher Hendrich 2 April 28, 206 Abstract. The aim of this article is to present
More informationLearning with stochastic proximal gradient
Learning with stochastic proximal gradient Lorenzo Rosasco DIBRIS, Università di Genova Via Dodecaneso, 35 16146 Genova, Italy lrosasco@mit.edu Silvia Villa, Băng Công Vũ Laboratory for Computational and
More informationBORIS MORDUKHOVICH Wayne State University Detroit, MI 48202, USA. Talk given at the SPCOM Adelaide, Australia, February 2015
CODERIVATIVE CHARACTERIZATIONS OF MAXIMAL MONOTONICITY BORIS MORDUKHOVICH Wayne State University Detroit, MI 48202, USA Talk given at the SPCOM 2015 Adelaide, Australia, February 2015 Based on joint papers
More information2 Regularized Image Reconstruction for Compressive Imaging and Beyond
EE 367 / CS 448I Computational Imaging and Display Notes: Compressive Imaging and Regularized Image Reconstruction (lecture ) Gordon Wetzstein gordon.wetzstein@stanford.edu This document serves as a supplement
More informationThe Subdifferential of Convex Deviation Measures and Risk Functions
The Subdifferential of Convex Deviation Measures and Risk Functions Nicole Lorenz Gert Wanka In this paper we give subdifferential formulas of some convex deviation measures using their conjugate functions
More informationEnhanced Fritz John Optimality Conditions and Sensitivity Analysis
Enhanced Fritz John Optimality Conditions and Sensitivity Analysis Dimitri P. Bertsekas Laboratory for Information and Decision Systems Massachusetts Institute of Technology March 2016 1 / 27 Constrained
More informationA New Primal Dual Algorithm for Minimizing the Sum of Three Functions with a Linear Operator
https://doi.org/10.1007/s10915-018-0680-3 A New Primal Dual Algorithm for Minimizing the Sum of Three Functions with a Linear Operator Ming Yan 1,2 Received: 22 January 2018 / Accepted: 22 February 2018
More informationRobust Duality in Parametric Convex Optimization
Robust Duality in Parametric Convex Optimization R.I. Boţ V. Jeyakumar G.Y. Li Revised Version: June 20, 2012 Abstract Modelling of convex optimization in the face of data uncertainty often gives rise
More informationGeneralized greedy algorithms.
Generalized greedy algorithms. François-Xavier Dupé & Sandrine Anthoine LIF & I2M Aix-Marseille Université - CNRS - Ecole Centrale Marseille, Marseille ANR Greta Séminaire Parisien des Mathématiques Appliquées
More informationarxiv: v1 [math.oc] 16 May 2018
An Algorithmic Framewor of Variable Metric Over-Relaxed Hybrid Proximal Extra-Gradient Method Li Shen 1 Peng Sun 1 Yitong Wang 1 Wei Liu 1 Tong Zhang 1 arxiv:180506137v1 [mathoc] 16 May 2018 Abstract We
More informationarxiv: v2 [cs.cv] 12 Sep 2014
Noname manuscript No. (will be inserted by the editor) An inertial forward-bacward algorithm for monotone inclusions D. Lorenz T. Poc arxiv:403.35v [cs.cv Sep 04 the date of receipt and acceptance should
More informationProximal methods. S. Villa. October 7, 2014
Proximal methods S. Villa October 7, 2014 1 Review of the basics Often machine learning problems require the solution of minimization problems. For instance, the ERM algorithm requires to solve a problem
More informationPIPA: A New Proximal Interior Point Algorithm for Large Scale Convex Optimization
PIPA: A New Proximal Interior Point Algorithm for Large Scale Convex Optimization Marie-Caroline Corbineau, Emilie Chouzenoux, Jean-Christophe Pesquet To cite this version: Marie-Caroline Corbineau, Emilie
More informationSemi-infinite programming, duality, discretization and optimality conditions
Semi-infinite programming, duality, discretization and optimality conditions Alexander Shapiro School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0205,
More informationAccelerated primal-dual methods for linearly constrained convex problems
Accelerated primal-dual methods for linearly constrained convex problems Yangyang Xu SIAM Conference on Optimization May 24, 2017 1 / 23 Accelerated proximal gradient For convex composite problem: minimize
More informationDual and primal-dual methods
ELE 538B: Large-Scale Optimization for Data Science Dual and primal-dual methods Yuxin Chen Princeton University, Spring 2018 Outline Dual proximal gradient method Primal-dual proximal gradient method
More informationarxiv: v1 [math.oc] 21 Apr 2016
Accelerated Douglas Rachford methods for the solution of convex-concave saddle-point problems Kristian Bredies Hongpeng Sun April, 06 arxiv:604.068v [math.oc] Apr 06 Abstract We study acceleration and
More informationA Globally Linearly Convergent Method for Pointwise Quadratically Supportable Convex-Concave Saddle Point Problems
A Globally Linearly Convergent Method for Pointwise Quadratically Supportable Convex-Concave Saddle Point Problems D. Russell Luke and Ron Shefi February 28, 2017 Abstract We study the Proximal Alternating
More informationarxiv: v1 [math.oc] 13 Dec 2018
A NEW HOMOTOPY PROXIMAL VARIABLE-METRIC FRAMEWORK FOR COMPOSITE CONVEX MINIMIZATION QUOC TRAN-DINH, LIANG LING, AND KIM-CHUAN TOH arxiv:8205243v [mathoc] 3 Dec 208 Abstract This paper suggests two novel
More informationAn inertial forward-backward algorithm for the minimization of the sum of two nonconvex functions
An inertial forward-backward algorithm for the minimization of the sum of two nonconvex functions Radu Ioan Boţ Ernö Robert Csetnek Szilárd Csaba László October, 1 Abstract. We propose a forward-backward
More informationMaster 2 MathBigData. 3 novembre CMAP - Ecole Polytechnique
Master 2 MathBigData S. Gaïffas 1 3 novembre 2014 1 CMAP - Ecole Polytechnique 1 Supervised learning recap Introduction Loss functions, linearity 2 Penalization Introduction Ridge Sparsity Lasso 3 Some
More informationShiqian Ma, MAT-258A: Numerical Optimization 1. Chapter 4. Subgradient
Shiqian Ma, MAT-258A: Numerical Optimization 1 Chapter 4 Subgradient Shiqian Ma, MAT-258A: Numerical Optimization 2 4.1. Subgradients definition subgradient calculus duality and optimality conditions Shiqian
More informationOn the Brézis - Haraux - type approximation in nonreflexive Banach spaces
On the Brézis - Haraux - type approximation in nonreflexive Banach spaces Radu Ioan Boţ Sorin - Mihai Grad Gert Wanka Abstract. We give Brézis - Haraux - type approximation results for the range of the
More informationAn Algorithmic Framework of Variable Metric Over-Relaxed Hybrid Proximal Extra-Gradient Method
An Algorithmic Framewor of Variable Metric Over-Relaxed Hybrid Proximal Extra-Gradient Method Li Shen 1 Peng Sun 1 Yitong Wang 1 Wei Liu 1 Tong Zhang 1 Abstract We propose a novel algorithmic framewor
More informationConvex Optimization Conjugate, Subdifferential, Proximation
1 Lecture Notes, HCI, 3.11.211 Chapter 6 Convex Optimization Conjugate, Subdifferential, Proximation Bastian Goldlücke Computer Vision Group Technical University of Munich 2 Bastian Goldlücke Overview
More informationIn Progress: Summary of Notation and Basic Results Convex Analysis C&O 663, Fall 2009
In Progress: Summary of Notation and Basic Results Convex Analysis C&O 663, Fall 2009 Intructor: Henry Wolkowicz November 26, 2009 University of Waterloo Department of Combinatorics & Optimization Abstract
More informationNonnegative Tensor Factorization using a proximal algorithm: application to 3D fluorescence spectroscopy
Nonnegative Tensor Factorization using a proximal algorithm: application to 3D fluorescence spectroscopy Caroline Chaux Joint work with X. Vu, N. Thirion-Moreau and S. Maire (LSIS, Toulon) Aix-Marseille
More informationPARALLEL SUBGRADIENT METHOD FOR NONSMOOTH CONVEX OPTIMIZATION WITH A SIMPLE CONSTRAINT
Linear and Nonlinear Analysis Volume 1, Number 1, 2015, 1 PARALLEL SUBGRADIENT METHOD FOR NONSMOOTH CONVEX OPTIMIZATION WITH A SIMPLE CONSTRAINT KAZUHIRO HISHINUMA AND HIDEAKI IIDUKA Abstract. In this
More informationDual Proximal Gradient Method
Dual Proximal Gradient Method http://bicmr.pku.edu.cn/~wenzw/opt-2016-fall.html Acknowledgement: this slides is based on Prof. Lieven Vandenberghes lecture notes Outline 2/19 1 proximal gradient method
More informationarxiv: v1 [math.na] 3 Jan 2019
arxiv manuscript No. (will be inserted by the editor) A Finite Element Nonoverlapping Domain Decomposition Method with Lagrange Multipliers for the Dual Total Variation Minimizations Chang-Ock Lee Jongho
More informationGSOS: Gauss-Seidel Operator Splitting Algorithm for Multi-Term Nonsmooth Convex Composite Optimization
: Gauss-Seidel Operator Splitting Algorithm for Multi-Term Nonsmooth Convex Composite Optimization Li Shen 1 Wei Liu 1 Ganzhao Yuan Shiqian Ma 3 Abstract In this paper, we propose a fast Gauss-Seidel Operator
More informationALGORITHMS FOR MINIMIZING DIFFERENCES OF CONVEX FUNCTIONS AND APPLICATIONS
ALGORITHMS FOR MINIMIZING DIFFERENCES OF CONVEX FUNCTIONS AND APPLICATIONS Mau Nam Nguyen (joint work with D. Giles and R. B. Rector) Fariborz Maseeh Department of Mathematics and Statistics Portland State
More informationFirst-order methods for structured nonsmooth optimization
First-order methods for structured nonsmooth optimization Sangwoon Yun Department of Mathematics Education Sungkyunkwan University Oct 19, 2016 Center for Mathematical Analysis & Computation, Yonsei University
More information9. Dual decomposition and dual algorithms
EE 546, Univ of Washington, Spring 2016 9. Dual decomposition and dual algorithms dual gradient ascent example: network rate control dual decomposition and the proximal gradient method examples with simple
More informationSPARSE SIGNAL RESTORATION. 1. Introduction
SPARSE SIGNAL RESTORATION IVAN W. SELESNICK 1. Introduction These notes describe an approach for the restoration of degraded signals using sparsity. This approach, which has become quite popular, is useful
More informationMath 273a: Optimization Subgradients of convex functions
Math 273a: Optimization Subgradients of convex functions Made by: Damek Davis Edited by Wotao Yin Department of Mathematics, UCLA Fall 2015 online discussions on piazza.com 1 / 42 Subgradients Assumptions
More informationAsynchronous Non-Convex Optimization For Separable Problem
Asynchronous Non-Convex Optimization For Separable Problem Sandeep Kumar and Ketan Rajawat Dept. of Electrical Engineering, IIT Kanpur Uttar Pradesh, India Distributed Optimization A general multi-agent
More informationOn the equivalence of the primal-dual hybrid gradient method and Douglas Rachford splitting
Mathematical Programming manuscript No. (will be inserted by the editor) On the equivalence of the primal-dual hybrid gradient method and Douglas Rachford splitting Daniel O Connor Lieven Vandenberghe
More informationMATH 680 Fall November 27, Homework 3
MATH 680 Fall 208 November 27, 208 Homework 3 This homework is due on December 9 at :59pm. Provide both pdf, R files. Make an individual R file with proper comments for each sub-problem. Subgradients and
More informationStrongly convex functions, Moreau envelopes and the generic nature of convex functions with strong minimizers
University of Wollongong Research Online Faculty of Engineering and Information Sciences - Papers: Part B Faculty of Engineering and Information Sciences 206 Strongly convex functions, Moreau envelopes
More informationLagrange duality. The Lagrangian. We consider an optimization program of the form
Lagrange duality Another way to arrive at the KKT conditions, and one which gives us some insight on solving constrained optimization problems, is through the Lagrange dual. The dual is a maximization
More informationA Coordinate Descent Primal-Dual Algorithm with Large Step Size and Possibly Non-Separable Functions
A Coordinate Descent Primal-Dual Algorithm with Large Step Size and Possibly Non-Separable Functions Olivier Fercoq Pascal Bianchi February 5, 2018 arxiv:1508.04625v2 [math.oc] 2 Feb 2018 Abstract This
More informationConvergence Rates for Projective Splitting
Convergence Rates for Projective Splitting Patric R. Johnstone Jonathan Ecstein August 8, 018 Abstract Projective splitting is a family of methods for solving inclusions involving sums of maximal monotone
More informationContraction Methods for Convex Optimization and Monotone Variational Inequalities No.16
XVI - 1 Contraction Methods for Convex Optimization and Monotone Variational Inequalities No.16 A slightly changed ADMM for convex optimization with three separable operators Bingsheng He Department of
More informationCoordinate Update Algorithm Short Course Subgradients and Subgradient Methods
Coordinate Update Algorithm Short Course Subgradients and Subgradient Methods Instructor: Wotao Yin (UCLA Math) Summer 2016 1 / 30 Notation f : H R { } is a closed proper convex function domf := {x R n
More informationOptimization for Learning and Big Data
Optimization for Learning and Big Data Donald Goldfarb Department of IEOR Columbia University Department of Mathematics Distinguished Lecture Series May 17-19, 2016. Lecture 1. First-Order Methods for
More informationFenchel-Moreau Conjugates of Inf-Transforms and Application to Stochastic Bellman Equation
Fenchel-Moreau Conjugates of Inf-Transforms and Application to Stochastic Bellman Equation Jean-Philippe Chancelier and Michel De Lara CERMICS, École des Ponts ParisTech First Conference on Discrete Optimization
More informationInertial Douglas-Rachford splitting for monotone inclusion problems
Inertial Douglas-Rachford splitting for monotone inclusion problems Radu Ioan Boţ Ernö Robert Csetnek Christopher Hendrich January 5, 2015 Abstract. We propose an inertial Douglas-Rachford splitting algorithm
More informationMAXIMALITY OF SUMS OF TWO MAXIMAL MONOTONE OPERATORS
MAXIMALITY OF SUMS OF TWO MAXIMAL MONOTONE OPERATORS JONATHAN M. BORWEIN, FRSC Abstract. We use methods from convex analysis convex, relying on an ingenious function of Simon Fitzpatrick, to prove maximality
More informationPrimal Solutions and Rate Analysis for Subgradient Methods
Primal Solutions and Rate Analysis for Subgradient Methods Asu Ozdaglar Joint work with Angelia Nedić, UIUC Conference on Information Sciences and Systems (CISS) March, 2008 Department of Electrical Engineering
More informationarxiv: v1 [math.oc] 29 Dec 2018
DeGroot-Friedkin Map in Opinion Dynamics is Mirror Descent Abhishek Halder ariv:1812.11293v1 [math.oc] 29 Dec 2018 Abstract We provide a variational interpretation of the DeGroot-Friedkin map in opinion
More informationSplitting methods for decomposing separable convex programs
Splitting methods for decomposing separable convex programs Philippe Mahey LIMOS - ISIMA - Université Blaise Pascal PGMO, ENSTA 2013 October 4, 2013 1 / 30 Plan 1 Max Monotone Operators Proximal techniques
More informationSEMI-SMOOTH SECOND-ORDER TYPE METHODS FOR COMPOSITE CONVEX PROGRAMS
SEMI-SMOOTH SECOND-ORDER TYPE METHODS FOR COMPOSITE CONVEX PROGRAMS XIANTAO XIAO, YONGFENG LI, ZAIWEN WEN, AND LIWEI ZHANG Abstract. The goal of this paper is to study approaches to bridge the gap between
More informationA generalized forward-backward method for solving split equality quasi inclusion problems in Banach spaces
Available online at www.isr-publications.com/jnsa J. Nonlinear Sci. Appl., 10 (2017), 4890 4900 Research Article Journal Homepage: www.tjnsa.com - www.isr-publications.com/jnsa A generalized forward-backward
More informationOptimization and Optimal Control in Banach Spaces
Optimization and Optimal Control in Banach Spaces Bernhard Schmitzer October 19, 2017 1 Convex non-smooth optimization with proximal operators Remark 1.1 (Motivation). Convex optimization: easier to solve,
More informationImage restoration: numerical optimisation
Image restoration: numerical optimisation Short and partial presentation Jean-François Giovannelli Groupe Signal Image Laboratoire de l Intégration du Matériau au Système Univ. Bordeaux CNRS BINP / 6 Context
More informationEE 367 / CS 448I Computational Imaging and Display Notes: Image Deconvolution (lecture 6)
EE 367 / CS 448I Computational Imaging and Display Notes: Image Deconvolution (lecture 6) Gordon Wetzstein gordon.wetzstein@stanford.edu This document serves as a supplement to the material discussed in
More information