Long-run Average Reward for Markov Decision Processes
|
|
- Winfred Mitchell
- 5 years ago
- Views:
Transcription
1 Long-run Average Reward for Markov Decision Processes Based on a paper at CAV 2017 Pranav Ashok 1, Krishnendu Chatterjee 2, Przemysław Daca 2, Jan Křetínský 1 and Tobias Meggendorfer 1 August 9, Technical University of Munich, Germany 2 IST Austria
2 Motivation Markov Decision Processes (MDPs) are a standard model for describing systems which display probabilistic as well as non-deterministic behaviour. 1
3 Motivation Open Problem Long-run Average Reward or Mean-payoff using Value Iteration (VI) VI approaches exist for subclasses, but not for general MDPs Other existing approaches: Linear Programming (LP) and Strategy Iteration (SI) 1 1 Martin L. Puterman, Markov Decision Processes,
4 Contributions Disprove conjectured stopping criterion for VI 2 General solution using VI Improve performance using ideas from Machine Learning 2 Not covered in this talk 3
5 Markov Decision Process (MDPs) b, 0 s 1 a, 10 a, 4 c, 0 s s 3 s 5 a, 10 a, 20 a, 0 a, 0 s 4 s 6 c, 0 4
6 Strategy A strategy (or policy) gives the action to be taken at every state. π := {s 1 b, s 2 a} b, 0 s 1 a, 10 a, 4 c, 0 s s 3 s 5 a, 10 a, 20 a, 0 a, 0 s 4 s 6 c, 0 5
7 Strategy A strategy (or policy) gives the action to be taken at every state. π := {s 1 a,..., s 6 a} b, 0 s 1 a, 10 a, 4 c, 0 s s 3 s 5 a, 10 a, 20 a, 0 a, 0 s 4 s 6 c, 0 5
8 Mean-payoff or Long-run Average Reward ρ = Then, n-step average reward is given by MP n (ρ) := 1 n n i=1 ρ i MP := sup π lim inf n Eπ [ MP n (ρ) ] 6
9 Value Iteration VI for Total Rewards (Bellman Equation) v n(s 1 ) = max { 0 + v n 1 (s 2 ), 10 + ( 0.99 v n 1 (s 3 ) v n 1 (s 5 ) ) } b, 0 s 1 a, 10 s 2 v n 1 (s 2 ) s 3 s 5 v n 1 (s 3 ) v n 1 (s 5 ) 7
10 Value Iteration Average reward 3 v n (s) lim n n v n (s) v n 1 (s) 3 After making the MDP aperiodic 8
11 Towards a General VI: Maximal End Components (MECs) s 3 s 3 s 4 s 4 An example MEC Communicating MDP VI 9
12 Towards a General VI: Step 1 MEC-Decomposition b, 0 s 1 a, 10 a, 4 s 2 c, 0 s 3 s 5 a, 10 a, 20 a, 0 a, 0 s 4 s 6 c, 0 Find all MECs 10
13 Towards a General VI: Step 1 MEC-Decomposition b, 0 s 1 a, 10 a, 4 4 c, a, 10 a, 20 a, 0 a, c, 0 Find all MECs and run VI on them until ε-convergence 10
14 Towards a General VI: Step 2 Weighted Reachability Max. mean-payoff = sup p p p 3 10 π p 1 s 1 p 3 4 p
15 Towards a General VI: Step 2 Weighted Reachability Max. mean-payoff R 4 = sup p 1 π R + p 2 5 R + p 3 10 R p 1 s 1 p R 5 R + p R 11
16 Towards a General VI: Step 2 Weighted Reachability Max. mean-payoff R 4 = sup p 1 π R + p 2 5 R + p 3 10 R p 1 s 1 p R 5 R + p R Mean-payoff reduced to reachability: P max ( +) 11
17 Limitations of this method Whole state space is being explored Finding MECs on large state spaces is computationally expensive 12
18 Improvement: avoid full state-space exploration Idea: let sampling guide us to the important regions b, 0 s1 a, 10 a, 4 c, 0 s s3 s5 a, 10 a, 20 a, 0 a, 0 s4 s6 c, 0 Contribution of red region to MP is potentially low 13
19 Improvement: guarantees through sampling Existing: BRTDP approach for reachability 4,5 Repeatedly samples paths from initial state Back-propagates values along the path using VI operator 4 Bounded Real-Time Dynamic Programming, McMahan et. al., ICML 05 5 Verification of Markov Decision Processes using Learning Algorithms, Brazdil et. al., ATVA 14 14
20 Improvement: BRTDP in action [0, 0] [0, 1] +[1, 1] 15
21 Improvement: BRTDP in action [0, 0] [0.4, 1] +[1, 1] 15
22 Improvement: BRTDP in action [0, 0] [0.4, 1] +[1, 1] 15
23 Improvement: BRTDP in action [0, 0] [0.4, 0.8] +[1, 1] 15
24 Improvement: BRTDP in action [0, 0] [0.4, 0.8] +[1, 1] 15
25 Improvement: BRTDP in action [0, 0] [0.4, 0.6] +[1, 1] 15
26 Improvement: BRTDP in action [0, 0] [0.4, 0.6] +[1, 1] 15
27 Improvement: BRTDP in action [0, 0] [5, 0.6] +[1, 1] 15
28 Improvement: Algorithm 1. Run BRTDP in search of the + state 2. Repeatedly collapse MECs amongst the states explored so far 3. While collapsing, compute their values and add special transition to + and states 16
29 Improvement: can do even more
30 Improvement: can do even more R 1 15 R R? Collapse MEC into a single state and add a special action 17
31 Final Algorithm: On-demand Value Iteration (ODV) 1. Sample paths like in BRTDP 2. When MECs are detected, collapse them but 3. Don t compute MEC value until ε-convergence 4. Add transition with probability (U-L) to? state 5. Refine value of MEC only when? encountered 18
32 Summary We saw two methods which can be used to obtain mean-payoff with guarantees 1. Collapse MECs, add transitions to +/ states, run reachability 2. ODV: Run sampling, collapse on-the-go, refine MEC values on-demand 19
33 Benchmarks Model States MECs LP 1 MEC-VI 2 virus cs_nfail investor phil-nofair TO 6.67 rabin TO MultiGain, Brazdil et. al MEC-VI: Straightforward conversion to reachability, then VI 20
34 Benchmarks On-demand VI better by orders of magnitude depending on topology Model States MEC-VI ODV ODV States ODV MECs zeroconf(40,10) MO avoid zeroconf(300,15) MO avoid sensors(2) sensors(3)
35 22
Infinite-Horizon Average Reward Markov Decision Processes
Infinite-Horizon Average Reward Markov Decision Processes Dan Zhang Leeds School of Business University of Colorado at Boulder Dan Zhang, Spring 2012 Infinite Horizon Average Reward MDP 1 Outline The average
More informationCSE250A Fall 12: Discussion Week 9
CSE250A Fall 12: Discussion Week 9 Aditya Menon (akmenon@ucsd.edu) December 4, 2012 1 Schedule for today Recap of Markov Decision Processes. Examples: slot machines and maze traversal. Planning and learning.
More informationMS&E338 Reinforcement Learning Lecture 1 - April 2, Introduction
MS&E338 Reinforcement Learning Lecture 1 - April 2, 2018 Introduction Lecturer: Ben Van Roy Scribe: Gabriel Maher 1 Reinforcement Learning Introduction In reinforcement learning (RL) we consider an agent
More informationProbabilistic Planning. George Konidaris
Probabilistic Planning George Konidaris gdk@cs.brown.edu Fall 2017 The Planning Problem Finding a sequence of actions to achieve some goal. Plans It s great when a plan just works but the world doesn t
More informationControlling probabilistic systems under partial observation an automata and verification perspective
Controlling probabilistic systems under partial observation an automata and verification perspective Nathalie Bertrand, Inria Rennes, France Uncertainty in Computation Workshop October 4th 2016, Simons
More informationReductions Of Undiscounted Markov Decision Processes and Stochastic Games To Discounted Ones. Jefferson Huang
Reductions Of Undiscounted Markov Decision Processes and Stochastic Games To Discounted Ones Jefferson Huang School of Operations Research and Information Engineering Cornell University November 16, 2016
More informationPreference Elicitation for Sequential Decision Problems
Preference Elicitation for Sequential Decision Problems Kevin Regan University of Toronto Introduction 2 Motivation Focus: Computational approaches to sequential decision making under uncertainty These
More informationStochastic Games with Time The value Min strategies Max strategies Determinacy Finite-state games Cont.-time Markov chains
Games with Time Finite-state Masaryk University Brno GASICS 00 /39 Outline Finite-state stochastic processes. Games over event-driven stochastic processes. Strategies,, determinacy. Existing results for
More informationReinforcement Learning
Reinforcement Learning Markov decision process & Dynamic programming Evaluative feedback, value function, Bellman equation, optimality, Markov property, Markov decision process, dynamic programming, value
More informationCSE 573. Markov Decision Processes: Heuristic Search & Real-Time Dynamic Programming. Slides adapted from Andrey Kolobov and Mausam
CSE 573 Markov Decision Processes: Heuristic Search & Real-Time Dynamic Programming Slides adapted from Andrey Kolobov and Mausam 1 Stochastic Shortest-Path MDPs: Motivation Assume the agent pays cost
More informationOptimism in the Face of Uncertainty Should be Refutable
Optimism in the Face of Uncertainty Should be Refutable Ronald ORTNER Montanuniversität Leoben Department Mathematik und Informationstechnolgie Franz-Josef-Strasse 18, 8700 Leoben, Austria, Phone number:
More informationPlanning in Markov Decision Processes
Carnegie Mellon School of Computer Science Deep Reinforcement Learning and Control Planning in Markov Decision Processes Lecture 3, CMU 10703 Katerina Fragkiadaki Markov Decision Process (MDP) A Markov
More informationMarkov Decision Processes
Markov Decision Processes Noel Welsh 11 November 2010 Noel Welsh () Markov Decision Processes 11 November 2010 1 / 30 Annoucements Applicant visitor day seeks robot demonstrators for exciting half hour
More informationSFM-11:CONNECT Summer School, Bertinoro, June 2011
SFM-:CONNECT Summer School, Bertinoro, June 20 EU-FP7: CONNECT LSCITS/PSS VERIWARE Part 3 Markov decision processes Overview Lectures and 2: Introduction 2 Discrete-time Markov chains 3 Markov decision
More informationThe Complexity of Ergodic Mean-payoff Games,
The Complexity of Ergodic Mean-payoff Games, Krishnendu Chatterjee Rasmus Ibsen-Jensen Abstract We study two-player (zero-sum) concurrent mean-payoff games played on a finite-state graph. We focus on the
More informationLecture notes for Analysis of Algorithms : Markov decision processes
Lecture notes for Analysis of Algorithms : Markov decision processes Lecturer: Thomas Dueholm Hansen June 6, 013 Abstract We give an introduction to infinite-horizon Markov decision processes (MDPs) with
More informationA Survey of Partial-Observation Stochastic Parity Games
Noname manuscript No. (will be inserted by the editor) A Survey of Partial-Observation Stochastic Parity Games Krishnendu Chatterjee Laurent Doyen Thomas A. Henzinger the date of receipt and acceptance
More informationAM 121: Intro to Optimization Models and Methods: Fall 2018
AM 11: Intro to Optimization Models and Methods: Fall 018 Lecture 18: Markov Decision Processes Yiling Chen Lesson Plan Markov decision processes Policies and value functions Solving: average reward, discounted
More informationMarkov Decision Processes Infinite Horizon Problems
Markov Decision Processes Infinite Horizon Problems Alan Fern * * Based in part on slides by Craig Boutilier and Daniel Weld 1 What is a solution to an MDP? MDP Planning Problem: Input: an MDP (S,A,R,T)
More informationReinforcement Learning
Reinforcement Learning Dipendra Misra Cornell University dkm@cs.cornell.edu https://dipendramisra.wordpress.com/ Task Grasp the green cup. Output: Sequence of controller actions Setup from Lenz et. al.
More information, and rewards and transition matrices as shown below:
CSE 50a. Assignment 7 Out: Tue Nov Due: Thu Dec Reading: Sutton & Barto, Chapters -. 7. Policy improvement Consider the Markov decision process (MDP) with two states s {0, }, two actions a {0, }, discount
More informationReinforcement Learning. Introduction
Reinforcement Learning Introduction Reinforcement Learning Agent interacts and learns from a stochastic environment Science of sequential decision making Many faces of reinforcement learning Optimal control
More informationPlanning Under Uncertainty II
Planning Under Uncertainty II Intelligent Robotics 2014/15 Bruno Lacerda Announcement No class next Monday - 17/11/2014 2 Previous Lecture Approach to cope with uncertainty on outcome of actions Markov
More information6 Reinforcement Learning
6 Reinforcement Learning As discussed above, a basic form of supervised learning is function approximation, relating input vectors to output vectors, or, more generally, finding density functions p(y,
More informationBasis Construction from Power Series Expansions of Value Functions
Basis Construction from Power Series Expansions of Value Functions Sridhar Mahadevan Department of Computer Science University of Massachusetts Amherst, MA 3 mahadeva@cs.umass.edu Bo Liu Department of
More informationA note on two-person zero-sum communicating stochastic games
Operations Research Letters 34 26) 412 42 Operations Research Letters www.elsevier.com/locate/orl A note on two-person zero-sum communicating stochastic games Zeynep Müge Avşar a, Melike Baykal-Gürsoy
More informationMultiagent Value Iteration in Markov Games
Multiagent Value Iteration in Markov Games Amy Greenwald Brown University with Michael Littman and Martin Zinkevich Stony Brook Game Theory Festival July 21, 2005 Agenda Theorem Value iteration converges
More informationPlanning and Model Selection in Data Driven Markov models
Planning and Model Selection in Data Driven Markov models Shie Mannor Department of Electrical Engineering Technion Joint work with many people along the way: Dotan Di-Castro (Yahoo!), Assaf Halak (Technion),
More information1 MDP Value Iteration Algorithm
CS 0. - Active Learning Problem Set Handed out: 4 Jan 009 Due: 9 Jan 009 MDP Value Iteration Algorithm. Implement the value iteration algorithm given in the lecture. That is, solve Bellman s equation using
More informationA subexponential lower bound for the Random Facet algorithm for Parity Games
A subexponential lower bound for the Random Facet algorithm for Parity Games Oliver Friedmann 1 Thomas Dueholm Hansen 2 Uri Zwick 3 1 Department of Computer Science, University of Munich, Germany. 2 Center
More informationMarkov Decision Processes Chapter 17. Mausam
Markov Decision Processes Chapter 17 Mausam Planning Agent Static vs. Dynamic Fully vs. Partially Observable Environment What action next? Deterministic vs. Stochastic Perfect vs. Noisy Instantaneous vs.
More informationInfinite-Horizon Discounted Markov Decision Processes
Infinite-Horizon Discounted Markov Decision Processes Dan Zhang Leeds School of Business University of Colorado at Boulder Dan Zhang, Spring 2012 Infinite Horizon Discounted MDP 1 Outline The expected
More informationMDP Preliminaries. Nan Jiang. February 10, 2019
MDP Preliminaries Nan Jiang February 10, 2019 1 Markov Decision Processes In reinforcement learning, the interactions between the agent and the environment are often described by a Markov Decision Process
More informationSection Notes 9. Midterm 2 Review. Applied Math / Engineering Sciences 121. Week of December 3, 2018
Section Notes 9 Midterm 2 Review Applied Math / Engineering Sciences 121 Week of December 3, 2018 The following list of topics is an overview of the material that was covered in the lectures and sections
More informationCMU Lecture 11: Markov Decision Processes II. Teacher: Gianni A. Di Caro
CMU 15-781 Lecture 11: Markov Decision Processes II Teacher: Gianni A. Di Caro RECAP: DEFINING MDPS Markov decision processes: o Set of states S o Start state s 0 o Set of actions A o Transitions P(s s,a)
More informationMachine Learning and Bayesian Inference. Unsupervised learning. Can we find regularity in data without the aid of labels?
Machine Learning and Bayesian Inference Dr Sean Holden Computer Laboratory, Room FC6 Telephone extension 6372 Email: sbh11@cl.cam.ac.uk www.cl.cam.ac.uk/ sbh11/ Unsupervised learning Can we find regularity
More informationValue and Policy Iteration
Chapter 7 Value and Policy Iteration 1 For infinite horizon problems, we need to replace our basic computational tool, the DP algorithm, which we used to compute the optimal cost and policy for finite
More informationChristopher Watkins and Peter Dayan. Noga Zaslavsky. The Hebrew University of Jerusalem Advanced Seminar in Deep Learning (67679) November 1, 2015
Q-Learning Christopher Watkins and Peter Dayan Noga Zaslavsky The Hebrew University of Jerusalem Advanced Seminar in Deep Learning (67679) November 1, 2015 Noga Zaslavsky Q-Learning (Watkins & Dayan, 1992)
More informationDecision Theory: Q-Learning
Decision Theory: Q-Learning CPSC 322 Decision Theory 5 Textbook 12.5 Decision Theory: Q-Learning CPSC 322 Decision Theory 5, Slide 1 Lecture Overview 1 Recap 2 Asynchronous Value Iteration 3 Q-Learning
More informationTwo Views on Multiple Mean-Payoff Objectives in Markov Decision Processes
wo Views on Multiple Mean-Payoff Objectives in Markov Decision Processes omáš Brázdil Faculty of Informatics Masaryk University Brno, Czech Republic brazdil@fi.muni.cz Václav Brožek LFCS, School of Informatics
More informationThe Simplex and Policy Iteration Methods are Strongly Polynomial for the Markov Decision Problem with Fixed Discount
The Simplex and Policy Iteration Methods are Strongly Polynomial for the Markov Decision Problem with Fixed Discount Yinyu Ye Department of Management Science and Engineering and Institute of Computational
More informationLecture 3: Markov Decision Processes
Lecture 3: Markov Decision Processes Joseph Modayil 1 Markov Processes 2 Markov Reward Processes 3 Markov Decision Processes 4 Extensions to MDPs Markov Processes Introduction Introduction to MDPs Markov
More informationMarkov decision processes and interval Markov chains: exploiting the connection
Markov decision processes and interval Markov chains: exploiting the connection Mingmei Teo Supervisors: Prof. Nigel Bean, Dr Joshua Ross University of Adelaide July 10, 2013 Intervals and interval arithmetic
More informationA Review of the E 3 Algorithm: Near-Optimal Reinforcement Learning in Polynomial Time
A Review of the E 3 Algorithm: Near-Optimal Reinforcement Learning in Polynomial Time April 16, 2016 Abstract In this exposition we study the E 3 algorithm proposed by Kearns and Singh for reinforcement
More informationDiscrete planning (an introduction)
Sistemi Intelligenti Corso di Laurea in Informatica, A.A. 2017-2018 Università degli Studi di Milano Discrete planning (an introduction) Nicola Basilico Dipartimento di Informatica Via Comelico 39/41-20135
More informationAn Adaptive Clustering Method for Model-free Reinforcement Learning
An Adaptive Clustering Method for Model-free Reinforcement Learning Andreas Matt and Georg Regensburger Institute of Mathematics University of Innsbruck, Austria {andreas.matt, georg.regensburger}@uibk.ac.at
More informationLearning to Coordinate Efficiently: A Model-based Approach
Journal of Artificial Intelligence Research 19 (2003) 11-23 Submitted 10/02; published 7/03 Learning to Coordinate Efficiently: A Model-based Approach Ronen I. Brafman Computer Science Department Ben-Gurion
More informationA Framework for Automated Competitive Analysis of On-line Scheduling of Firm-Deadline Tasks
A Framework for Automated Competitive Analysis of On-line Scheduling of Firm-Deadline Tasks Krishnendu Chatterjee 1, Andreas Pavlogiannis 1, Alexander Kößler 2, Ulrich Schmid 2 1 IST Austria, 2 TU Wien
More informationOptimally Solving Dec-POMDPs as Continuous-State MDPs
Optimally Solving Dec-POMDPs as Continuous-State MDPs Jilles Dibangoye (1), Chris Amato (2), Olivier Buffet (1) and François Charpillet (1) (1) Inria, Université de Lorraine France (2) MIT, CSAIL USA IJCAI
More informationPrioritized Sweeping Converges to the Optimal Value Function
Technical Report DCS-TR-631 Prioritized Sweeping Converges to the Optimal Value Function Lihong Li and Michael L. Littman {lihong,mlittman}@cs.rutgers.edu RL 3 Laboratory Department of Computer Science
More informationPART A and ONE question from PART B; or ONE question from PART A and TWO questions from PART B.
Advanced Topics in Machine Learning, GI13, 2010/11 Advanced Topics in Machine Learning, GI13, 2010/11 Answer any THREE questions. Each question is worth 20 marks. Use separate answer books Answer any THREE
More informationMarkov Decision Processes and Dynamic Programming
Markov Decision Processes and Dynamic Programming A. LAZARIC (SequeL Team @INRIA-Lille) Ecole Centrale - Option DAD SequeL INRIA Lille EC-RL Course In This Lecture A. LAZARIC Markov Decision Processes
More informationNovember 28 th, Carlos Guestrin 1. Lower dimensional projections
PCA Machine Learning 10701/15781 Carlos Guestrin Carnegie Mellon University November 28 th, 2007 1 Lower dimensional projections Rather than picking a subset of the features, we can new features that are
More informationLecture 3: Policy Evaluation Without Knowing How the World Works / Model Free Policy Evaluation
Lecture 3: Policy Evaluation Without Knowing How the World Works / Model Free Policy Evaluation CS234: RL Emma Brunskill Winter 2018 Material builds on structure from David SIlver s Lecture 4: Model-Free
More informationExpectations or Guarantees? I Want It All! A Crossroad between Games and MDPs
Expectations or Guarantees? I Want It All! A Crossroad between Games and MDPs V. Bruyère (UMONS) E. Filiot (ULB) M. Randour (UMONS-ULB) J.-F. Raskin (ULB) Grenoble - 0.04.04 SR 04 - nd International Workshop
More informationInternet Monetization
Internet Monetization March May, 2013 Discrete time Finite A decision process (MDP) is reward process with decisions. It models an environment in which all states are and time is divided into stages. Definition
More informationMarkov Decision Processes Chapter 17. Mausam
Markov Decision Processes Chapter 17 Mausam Planning Agent Static vs. Dynamic Fully vs. Partially Observable Environment What action next? Deterministic vs. Stochastic Perfect vs. Noisy Instantaneous vs.
More informationUnderstanding (Exact) Dynamic Programming through Bellman Operators
Understanding (Exact) Dynamic Programming through Bellman Operators Ashwin Rao ICME, Stanford University January 15, 2019 Ashwin Rao (Stanford) Bellman Operators January 15, 2019 1 / 11 Overview 1 Value
More informationRandomness for Free. 1 Introduction. Krishnendu Chatterjee 1, Laurent Doyen 2, Hugo Gimbert 3, and Thomas A. Henzinger 1
Randomness for Free Krishnendu Chatterjee 1, Laurent Doyen 2, Hugo Gimbert 3, and Thomas A. Henzinger 1 1 IST Austria (Institute of Science and Technology Austria) 2 LSV, ENS Cachan & CNRS, France 3 LaBri
More informationCrowdsourcing & Optimal Budget Allocation in Crowd Labeling
Crowdsourcing & Optimal Budget Allocation in Crowd Labeling Madhav Mohandas, Richard Zhu, Vincent Zhuang May 5, 2016 Table of Contents 1. Intro to Crowdsourcing 2. The Problem 3. Knowledge Gradient Algorithm
More informationBasics of reinforcement learning
Basics of reinforcement learning Lucian Buşoniu TMLSS, 20 July 2018 Main idea of reinforcement learning (RL) Learn a sequential decision policy to optimize the cumulative performance of an unknown system
More informationReinforcement Learning. Spring 2018 Defining MDPs, Planning
Reinforcement Learning Spring 2018 Defining MDPs, Planning understandability 0 Slide 10 time You are here Markov Process Where you will go depends only on where you are Markov Process: Information state
More informationThis question has three parts, each of which can be answered concisely, but be prepared to explain and justify your concise answer.
This question has three parts, each of which can be answered concisely, but be prepared to explain and justify your concise answer. 1. Suppose you have a policy and its action-value function, q, then you
More informationPlanning by Probabilistic Inference
Planning by Probabilistic Inference Hagai Attias Microsoft Research 1 Microsoft Way Redmond, WA 98052 Abstract This paper presents and demonstrates a new approach to the problem of planning under uncertainty.
More informationChapter 16 Planning Based on Markov Decision Processes
Lecture slides for Automated Planning: Theory and Practice Chapter 16 Planning Based on Markov Decision Processes Dana S. Nau University of Maryland 12:48 PM February 29, 2012 1 Motivation c a b Until
More informationPART A and ONE question from PART B; or ONE question from PART A and TWO questions from PART B.
Advanced Topics in Machine Learning, GI13, 2010/11 Advanced Topics in Machine Learning, GI13, 2010/11 Answer any THREE questions. Each question is worth 20 marks. Use separate answer books Answer any THREE
More informationReinforcement Learning and Control
CS9 Lecture notes Andrew Ng Part XIII Reinforcement Learning and Control We now begin our study of reinforcement learning and adaptive control. In supervised learning, we saw algorithms that tried to make
More informationAn Empirical Algorithm for Relative Value Iteration for Average-cost MDPs
2015 IEEE 54th Annual Conference on Decision and Control CDC December 15-18, 2015. Osaka, Japan An Empirical Algorithm for Relative Value Iteration for Average-cost MDPs Abhishek Gupta Rahul Jain Peter
More informationMarkov Decision Processes (and a small amount of reinforcement learning)
Markov Decision Processes (and a small amount of reinforcement learning) Slides adapted from: Brian Williams, MIT Manuela Veloso, Andrew Moore, Reid Simmons, & Tom Mitchell, CMU Nicholas Roy 16.4/13 Session
More informationLinear Programming Methods
Chapter 11 Linear Programming Methods 1 In this chapter we consider the linear programming approach to dynamic programming. First, Bellman s equation can be reformulated as a linear program whose solution
More informationControl Theory : Course Summary
Control Theory : Course Summary Author: Joshua Volkmann Abstract There are a wide range of problems which involve making decisions over time in the face of uncertainty. Control theory draws from the fields
More informationMarkov decision processes
CS 2740 Knowledge representation Lecture 24 Markov decision processes Milos Hauskrecht milos@cs.pitt.edu 5329 Sennott Square Administrative announcements Final exam: Monday, December 8, 2008 In-class Only
More informationApproximate dynamic programming for stochastic reachability
Approximate dynamic programming for stochastic reachability Nikolaos Kariotoglou, Sean Summers, Tyler Summers, Maryam Kamgarpour and John Lygeros Abstract In this work we illustrate how approximate dynamic
More informationLecture 17: Reinforcement Learning, Finite Markov Decision Processes
CSE599i: Online and Adaptive Machine Learning Winter 2018 Lecture 17: Reinforcement Learning, Finite Markov Decision Processes Lecturer: Kevin Jamieson Scribes: Aida Amini, Kousuke Ariga, James Ferguson,
More informationProbabilistic model checking with PRISM
Probabilistic model checking with PRISM Marta Kwiatkowska Department of Computer Science, University of Oxford 4th SSFT, Menlo College, May 204 Part 2 Markov decision processes Overview (Part 2) Introduction
More informationReinforcement Learning. Yishay Mansour Tel-Aviv University
Reinforcement Learning Yishay Mansour Tel-Aviv University 1 Reinforcement Learning: Course Information Classes: Wednesday Lecture 10-13 Yishay Mansour Recitations:14-15/15-16 Eliya Nachmani Adam Polyak
More informationReinforcement Learning. George Konidaris
Reinforcement Learning George Konidaris gdk@cs.brown.edu Fall 2017 Machine Learning Subfield of AI concerned with learning from data. Broadly, using: Experience To Improve Performance On Some Task (Tom
More informationExponential Moving Average Based Multiagent Reinforcement Learning Algorithms
Exponential Moving Average Based Multiagent Reinforcement Learning Algorithms Mostafa D. Awheda Department of Systems and Computer Engineering Carleton University Ottawa, Canada KS 5B6 Email: mawheda@sce.carleton.ca
More informationArtificial Intelligence & Sequential Decision Problems
Artificial Intelligence & Sequential Decision Problems (CIV6540 - Machine Learning for Civil Engineers) Professor: James-A. Goulet Département des génies civil, géologique et des mines Chapter 15 Goulet
More informationMarkov Decision Processes with Multiple Long-run Average Objectives
Markov Decision Processes with Multiple Long-run Average Objectives Krishnendu Chatterjee UC Berkeley c krish@eecs.berkeley.edu Abstract. We consider Markov decision processes (MDPs) with multiple long-run
More informationCS 7180: Behavioral Modeling and Decisionmaking
CS 7180: Behavioral Modeling and Decisionmaking in AI Markov Decision Processes for Complex Decisionmaking Prof. Amy Sliva October 17, 2012 Decisions are nondeterministic In many situations, behavior and
More informationProbabilistic Model Checking and Strategy Synthesis for Robot Navigation
Probabilistic Model Checking and Strategy Synthesis for Robot Navigation Dave Parker University of Birmingham (joint work with Bruno Lacerda, Nick Hawes) AIMS CDT, Oxford, May 2015 Overview Probabilistic
More informationDecision Theory: Markov Decision Processes
Decision Theory: Markov Decision Processes CPSC 322 Lecture 33 March 31, 2006 Textbook 12.5 Decision Theory: Markov Decision Processes CPSC 322 Lecture 33, Slide 1 Lecture Overview Recap Rewards and Policies
More informationCourse 16:198:520: Introduction To Artificial Intelligence Lecture 13. Decision Making. Abdeslam Boularias. Wednesday, December 7, 2016
Course 16:198:520: Introduction To Artificial Intelligence Lecture 13 Decision Making Abdeslam Boularias Wednesday, December 7, 2016 1 / 45 Overview We consider probabilistic temporal models where the
More informationREINFORCEMENT LEARNING
REINFORCEMENT LEARNING Larry Page: Where s Google going next? DeepMind's DQN playing Breakout Contents Introduction to Reinforcement Learning Deep Q-Learning INTRODUCTION TO REINFORCEMENT LEARNING Contents
More informationQ-Learning for Markov Decision Processes*
McGill University ECSE 506: Term Project Q-Learning for Markov Decision Processes* Authors: Khoa Phan khoa.phan@mail.mcgill.ca Sandeep Manjanna sandeep.manjanna@mail.mcgill.ca (*Based on: Convergence of
More informationA Survey of Stochastic ω-regular Games
A Survey of Stochastic ω-regular Games Krishnendu Chatterjee Thomas A. Henzinger EECS, University of California, Berkeley, USA Computer and Communication Sciences, EPFL, Switzerland {c krish,tah}@eecs.berkeley.edu
More informationTemporal Difference Learning & Policy Iteration
Temporal Difference Learning & Policy Iteration Advanced Topics in Reinforcement Learning Seminar WS 15/16 ±0 ±0 +1 by Tobias Joppen 03.11.2015 Fachbereich Informatik Knowledge Engineering Group Prof.
More informationLaplacian Agent Learning: Representation Policy Iteration
Laplacian Agent Learning: Representation Policy Iteration Sridhar Mahadevan Example of a Markov Decision Process a1: $0 Heaven $1 Earth What should the agent do? a2: $100 Hell $-1 V a1 ( Earth ) = f(0,1,1,1,1,...)
More informationThe Markov Decision Process (MDP) model
Decision Making in Robots and Autonomous Agents The Markov Decision Process (MDP) model Subramanian Ramamoorthy School of Informatics 25 January, 2013 In the MAB Model We were in a single casino and the
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 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 informationReinforcement Learning as Classification Leveraging Modern Classifiers
Reinforcement Learning as Classification Leveraging Modern Classifiers Michail G. Lagoudakis and Ronald Parr Department of Computer Science Duke University Durham, NC 27708 Machine Learning Reductions
More informationLecture 1: March 7, 2018
Reinforcement Learning Spring Semester, 2017/8 Lecture 1: March 7, 2018 Lecturer: Yishay Mansour Scribe: ym DISCLAIMER: Based on Learning and Planning in Dynamical Systems by Shie Mannor c, all rights
More informationQ-Learning in Continuous State Action Spaces
Q-Learning in Continuous State Action Spaces Alex Irpan alexirpan@berkeley.edu December 5, 2015 Contents 1 Introduction 1 2 Background 1 3 Q-Learning 2 4 Q-Learning In Continuous Spaces 4 5 Experimental
More informationReinforcement Learning
Reinforcement Learning Model-Based Reinforcement Learning Model-based, PAC-MDP, sample complexity, exploration/exploitation, RMAX, E3, Bayes-optimal, Bayesian RL, model learning Vien Ngo MLR, University
More informationReinforcement Learning. Machine Learning, Fall 2010
Reinforcement Learning Machine Learning, Fall 2010 1 Administrativia This week: finish RL, most likely start graphical models LA2: due on Thursday LA3: comes out on Thursday TA Office hours: Today 1:30-2:30
More informationMARKOV DECISION PROCESSES (MDP) AND REINFORCEMENT LEARNING (RL) Versione originale delle slide fornita dal Prof. Francesco Lo Presti
1 MARKOV DECISION PROCESSES (MDP) AND REINFORCEMENT LEARNING (RL) Versione originale delle slide fornita dal Prof. Francesco Lo Presti Historical background 2 Original motivation: animal learning Early
More informationRobust Modified Policy Iteration
Robust Modified Policy Iteration David L. Kaufman Department of Industrial and Operations Engineering, University of Michigan 1205 Beal Avenue, Ann Arbor, MI 48109, USA 8davidlk8umich.edu (remove 8s) Andrew
More information