Dynamic Programming and Reinforcement Learning
|
|
- Harry Emil Bryan
- 5 years ago
- Views:
Transcription
1 Dynamic Programming and Reinforcement Learning Daniel Russo Columbia Business School Decision Risk and Operations Division Fall, 2017 Daniel Russo (Columbia) Fall / 34
2 Supervised Machine Learning Learning from datasets A passive paradigm Focus on pattern recognition Daniel Russo (Columbia) Fall / 34
3 Reinforcement Learning Reward Action Outcome Environment Learning to attain a goal through interaction with a poorly understood environment. Daniel Russo (Columbia) Fall / 34
4 Canonical (and toy) RL environments Cart Pole Mountain Car Daniel Russo (Columbia) Fall / 34
5 Impressive new (and toy) RL environments Atari from pixels Daniel Russo (Columbia) Fall / 34
6 Challenges in Reinforcement Learning Partial Feedback The data one gathers depends on the actions they take. Delayed Consequences Rather than maximize the immediate benefit from the next interaction, one must consider the impact on future interactions. Daniel Russo (Columbia) Fall / 34
7 Dream Application: Management of Chronic Diseases Various researchers are working on mobile health interventions Daniel Russo (Columbia) Fall / 34
8 Dream Application: Intelligent Tutoring Systems *Picture shamelessly lifted from a slide of Emma Brunskill s Daniel Russo (Columbia) Fall / 34
9 Dream Application: Beyond Myopia in E-Commerce Online marketplaces and web services have repeated interactions with users, but are deigned to optimize the next interaction. RL provides a framework for optimizing the cumulative value generated by such interactions. How useful will this turn out to be? Daniel Russo (Columbia) Fall / 34
10 Deep Reinforcement Learning RL where function approximation is performed using a deep neural network, instead of using linear models, kernel methods, shallow neural networks, etc. Daniel Russo (Columbia) Fall / 34
11 Deep Reinforcement Learning RL where function approximation is performed using a deep neural network, instead of using linear models, kernel methods, shallow neural networks, etc. Justified excitement Hope is to enable direct training of control systems based on complex sensory inputs (e.g. visual or auditory sensors) DeepMind s DQN learns to play Atari from pixels, without learning vision first. Daniel Russo (Columbia) Fall / 34
12 Deep Reinforcement Learning RL where function approximation is performed using a deep neural network, instead of using linear models, kernel methods, shallow neural networks, etc. Justified excitement Hope is to enable direct training of control systems based on complex sensory inputs (e.g. visual or auditory sensors) DeepMind s DQN learns to play Atari from pixels, without learning vision first. Also a lot of less justified hype. Daniel Russo (Columbia) Fall / 34
13 Warning 1 This is an advanced PhD course. 2 It will be primarily theoretical. We will prove theorems when we can. The emphasis will be on precise understand of why methods work and why they may fail completely in simple cases. 3 There are tons of engineering tricks to Deep RL. I won t cover these. Daniel Russo (Columbia) Fall / 34
14 My Goals 1 Encourage great students to do research in this area. 2 Provide a fun platform for introducing technical tools to operations PhD students. Dynamic programming, stochastic approximation, exploration algorithms and regret analysis. 3 Sharpen my own understanding. Daniel Russo (Columbia) Fall / 34
15 Tentative Course Outline 1 Foundational Material on MDPs 2 Estimating Long Run Value 3 Exploration Algorithms * Additional topics as time permits Policy gradients and actor critic Rollout and Monte-Carlo tree search. Daniel Russo (Columbia) Fall / 34
16 Markov Decision Processes: A warmup On the white-board Shortest path in a directed graph Daniel Russo (Columbia) Fall / 34
17 Markov Decision Processes: A warmup On the white-board Shortest path in a directed graph Imagine while traversing the shortest path, you discover one of the routes is closed. How should you adjust your path? Daniel Russo (Columbia) Fall / 34
18 Example: Inventory Control Stochastic demand Orders have lead time Non-perishable inventory Inventory holding costs Finite selling season Daniel Russo (Columbia) Fall / 34
19 Example: Inventory Control Periods k = 0, 1, 2,..., N x k {0,..., 1000} u k {0,..., 1000 x k } w k {0, 1, 2,...} x k+1 = x k w k + u k current inventory inventory order demand (i.i.d w/ known dist.) Transition dynamics Daniel Russo (Columbia) Fall / 34
20 Example: Inventory Control Periods k = 0, 1, 2,..., N x k {0,..., 1000} u k {0,..., 1000 x k } w k {0, 1, 2,...} x k+1 = x k w k + u k current inventory inventory order demand (i.i.d w/ known dist.) Transition dynamics Cost function g(x, u, w) = c H x }{{} Holding cost + c L w x } {{ } Lost sales + c O (u) }{{} Order cost Daniel Russo (Columbia) Fall / 34
21 Example: Inventory Control Objective: min E N k=0 g(x k, u k, w k ) Daniel Russo (Columbia) Fall / 34
22 Example: Inventory Control Objective: min E N k=0 Minimize over what? g(x k, u k, w k ) Daniel Russo (Columbia) Fall / 34
23 Example: Inventory Control Objective: min E N k=0 Minimize over what? g(x k, u k, w k ) Over fixed sequences of controls u0, u 1,...? No, over policies (adaptive ordering strategies). Daniel Russo (Columbia) Fall / 34
24 Example: Inventory Control Objective: min E N k=0 Minimize over what? g(x k, u k, w k ) Over fixed sequences of controls u0, u 1,...? No, over policies (adaptive ordering strategies). Sequential decision making under uncertainty where Decisions have delayed consequences. Relevant information is revealed during the decision process. Daniel Russo (Columbia) Fall / 34
25 Further Examples Dynamic pricing (over a selling season) Trade execution (with market impact) Queuing admission control Consumption-savings models in economics Search models in economics Timing of maintenance and repairs Daniel Russo (Columbia) Fall / 34
26 Finite Horizon MDPs: formulation A discrete time dynamic system where x k+1 = f k (x k, u k, w k ) x k X k u k U k (x k ) w k k = 0, 1,..., N state control (i.i.d w/ known dist.) Assume state and control spaces are finite. Daniel Russo (Columbia) Fall / 34
27 Finite Horizon MDPs: formulation A discrete time dynamic system where x k+1 = f k (x k, u k, w k ) x k X k u k U k (x k ) w k k = 0, 1,..., N state control (i.i.d w/ known dist.) Assume state and control spaces are finite. The total cost incurred is N k=0 g k (x k, u k, w k ) } {{ } cost in period k Daniel Russo (Columbia) Fall / 34
28 Finite Horizon MDPs: formulation A policy is a sequence π = (µ 0, µ 1,..., µ N ) where µ k : x k u k U k (x k ). Expected cost of following π from state x 0 is J π (x 0 ) = E N k=0 g k (x k, u k, w k ) where x k+1 = f k (x k, u k, w k ) and E[ ] is over the w ks. Daniel Russo (Columbia) Fall / 34
29 Finite Horizon MDPs: formulation The optimal expected cost to go from x 0 is J (x 0 ) = min π Π J π(x 0 ) where Π consists of all feasible policies. We will see the same policy π is optimal for all initial states. So J (x) = J π (x) x Daniel Russo (Columbia) Fall / 34
30 Minor differences with Bertsekas Vol. I Bertsekas Uses a special terminal cost function g N (x N ) Can always take gn (x, u, w) to be independent of u, w. Lets the distribution of w k depend on k and x k. This can be embedded in the functions f k, g k. Daniel Russo (Columbia) Fall / 34
31 Principle of Optimality Regardless of the consequences of initial decisions, an optimal policy should be optimal in the sub-problem beginning in the current state and time period. Daniel Russo (Columbia) Fall / 34
32 Principle of Optimality Regardless of the consequences of initial decisions, an optimal policy should be optimal in the sub-problem beginning in the current state and time period. Sufficiency: Such policies exist and minimize total expected cost from any initial state. Necessity: A policy that is optimal from some initial state must behave optimally in any subproblem that is reached with positive probability. Daniel Russo (Columbia) Fall / 34
33 The Dynamic Programming Algorithm Set J N(x) = min E[g N(x, u, w)] u U N (x) x X n For k = N 1, N 2,... 0, set J k (x) = min E[g k(x, u, w)+j u U k k+1(f k (x, u, w))] x X k. (x) Daniel Russo (Columbia) Fall / 34
34 The Dynamic Programming Algorithm Proposition For all x X 0, J (x) = J 0 (x). The optimal cost to go is attained by a policy π = (µ 0,..., µ N ) where µ N (x) arg min u U N (x) E[g N (x, u, w)] x X N and for all k {0,..., N 1}, x X k µ k(x) arg min u U k (x) E[g k(x, u, w) + J k+1(f k (x, u, w))]. Daniel Russo (Columbia) Fall / 34
35 The Dynamic Programming Algorithm Class Exercise Argue this is true for a 2 period problem (N=1). Hint, recall the tower property of conditional expectation. E[Y ] = E[E[Y X]] Daniel Russo (Columbia) Fall / 34
36 A Tedious Proof For any policy π = (µ 0, µ 1 ) and initial state x 0, E π [g 0 (x 0, µ 0 (x 0 ), w 0 ) + g 1 (x 1, µ 1 (x 1 ), w 1 )] = E π [g 0 (x 0, µ 0 (x 0 ), w 0 ) + E[g 1 (x 1, µ 1 (x 1 ), w 1 ) x 1 ]] E π [g 0 (x 0, µ 0 (x 0 ), w 0 ) + min E[g 1(x 1, u, w 1 ) x 1 ]] u U(x 1 ) = E π [g 0 (x 0, µ 0 (x 0 ), w 0 ) + J1 (x 1 )] = E π [g 0 (x 0, µ 0 (x 0 ), w 0 ) + J1 (f 0 (x 0, µ 0 (x 0 ), w 0 ))] min E[g 0(x 0, u, w 0 ) + J1 (f 0 (x 0, u, w 0 )] u U(x 0 ) = J0 (x 0 ) Under π, every inequality is an equality. Daniel Russo (Columbia) Fall / 34
37 Markov Property Markov Chain A stochastic process (X 0, X 1, X 2,...) is a Markov chain if for each n N, conditioned on X n 1, X n is independent of (X 0,..., X n 2 ). That is P(X n = X n 1 ) = P(X n = X 0,..., X n 1 ) Daniel Russo (Columbia) Fall / 34
38 Markov Property Markov Chain A stochastic process (X 0, X 1, X 2,...) is a Markov chain if for each n N, conditioned on X n 1, X n is independent of (X 0,..., X n 2 ). That is P(X n = X n 1 ) = P(X n = X 0,..., X n 1 ) Without loss of generality we can view a Markov chain as the output of a stochastic recursion X n+1 = f n (X n, W n ) for an i.i.d sequence of disturbances (W 0, W 1,...). Daniel Russo (Columbia) Fall / 34
39 Markov Property Our problem is called a Markov decision process because P(x n+1 = x x 0, u 0, w 0,..., x n, u n ) = P(f n (x n, u n, w n ) = x x n, u n ) = P(x n+1 = x x n, u n ) Requires the encoding of the state is sufficiently rich. Daniel Russo (Columbia) Fall / 34
40 Inventory Control Revisited Suppose that inventory has a lead time of 2 periods. Orders can still be placed in any period. Is this an MDP with state=current inventory? Daniel Russo (Columbia) Fall / 34
41 Inventory Control Revisited Suppose that inventory has a lead time of 2 periods. Orders can still be placed in any period. Is this an MDP with state=current inventory? No! Transition probabilities depend on the order that is currently in transit. Daniel Russo (Columbia) Fall / 34
42 Inventory Control Revisited Suppose that inventory has a lead time of 2 periods. Orders can still be placed in any period. Is this an MDP with state=current inventory? No! Transition probabilities depend on the order that is currently in transit. This is an MDP if we augment the state space so x n = (current inventory, inventory arriving next period). Daniel Russo (Columbia) Fall / 34
43 State Augmentation In the extreme, choosing the state to be the full history x n 1 = (x 0, u 0,..., u n 2, x n 1 ) suffices since P( x n = x n 1, u n 1 ) = P( x n = x 0, u 0,..., x n 1, u n 1 ). Daniel Russo (Columbia) Fall / 34
44 State Augmentation In the extreme, choosing the state to be the full history x n 1 = (x 0, u 0,..., u n 2, x n 1 ) suffices since P( x n = x n 1, u n 1 ) = P( x n = x 0, u 0,..., x n 1, u n 1 ). For the next few weeks we will assume the Markov property holds. Daniel Russo (Columbia) Fall / 34
45 State Augmentation In the extreme, choosing the state to be the full history x n 1 = (x 0, u 0,..., u n 2, x n 1 ) suffices since P( x n = x n 1, u n 1 ) = P( x n = x 0, u 0,..., x n 1, u n 1 ). For the next few weeks we will assume the Markov property holds. Computational tractability usually requires a compact state representation. Daniel Russo (Columbia) Fall / 34
46 Example: selling an asset An instance of optimal stopping. Deadline to sell within N periods. Potential buyers make offers in sequence. The agent chooses to accept or reject each offer The asset is sold once an offer is accepted. Offers are no longer available once declined. Offers are statistically independent. Profits can be invested with interest rate r > 0 per period. Daniel Russo (Columbia) Fall / 34
47 Example: selling an asset An instance of optimal stopping. Deadline to sell within N periods. Potential buyers make offers in sequence. The agent chooses to accept or reject each offer The asset is sold once an offer is accepted. Offers are no longer available once declined. Offers are statistically independent. Profits can be invested with interest rate r > 0 per period. Class Exercise 1 Formulate this as a finite horizon MDP. 2 Write down the DP algorithm. Daniel Russo (Columbia) Fall / 34
48 Example: selling an asset Special terminal state t (costless and absorbing) x k t is the offer considered at time k. x 0 = 0 is fictitious null offer. g k (x k, sell) = (1 + r) N k x k. x k = w k 1 for independent w 0, w 1,... Daniel Russo (Columbia) Fall / 34
49 Example: selling an asset DP Algorithm Jk (t) = 0 k JN(x) = x Jk (x) = max{(1 + r) N k x, E[Jk+1(w k )]} A threshold policy is optimal: Sell x k E[J k+1(w k )] (1 + r) N k Daniel Russo (Columbia) Fall / 34
Lecture 17: More on Markov Decision Processes. Reinforcement learning
Lecture 17: More on Markov Decision Processes. Reinforcement learning Learning a model: maximum likelihood Learning a value function directly Monte Carlo Temporal-difference (TD) learning COMP-424, Lecture
More informationStochastic Optimal Control
Stochastic Optimal Control Lecturer: Eilyan Bitar, Cornell ECE Scribe: Kevin Kircher, Cornell MAE These notes summarize some of the material from ECE 5555 (Stochastic Systems) at Cornell in the fall of
More informationDynamic Pricing with Varying Cost
Dynamic Pricing with Varying Cost L. Jeff Hong College of Business City University of Hong Kong Joint work with Ying Zhong and Guangwu Liu Outline 1 Introduction 2 Problem Formulation 3 Pricing Policy
More information6.231 DYNAMIC PROGRAMMING LECTURE 5 LECTURE OUTLINE
6.231 DYNAMIC PROGRAMMING LECTURE 5 LECTURE OUTLINE Stopping problems Scheduling problems Minimax Control 1 PURE STOPPING PROBLEMS Two possible controls: Stop (incur a one-time stopping cost, and move
More information6.231 DYNAMIC PROGRAMMING LECTURE 10 LECTURE OUTLINE
6.231 DYNAMIC PROGRAMMING LECTURE 10 LECTURE OUTLINE Rollout algorithms Cost improvement property Discrete deterministic problems Approximations of rollout algorithms Discretization of continuous time
More informationSequential Decision Making
Sequential Decision Making Dynamic programming Christos Dimitrakakis Intelligent Autonomous Systems, IvI, University of Amsterdam, The Netherlands March 18, 2008 Introduction Some examples Dynamic programming
More informationHandout 8: Introduction to Stochastic Dynamic Programming. 2 Examples of Stochastic Dynamic Programming Problems
SEEM 3470: Dynamic Optimization and Applications 2013 14 Second Term Handout 8: Introduction to Stochastic Dynamic Programming Instructor: Shiqian Ma March 10, 2014 Suggested Reading: Chapter 1 of Bertsekas,
More informationMonte-Carlo Planning: Introduction and Bandit Basics. Alan Fern
Monte-Carlo Planning: Introduction and Bandit Basics Alan Fern 1 Large Worlds We have considered basic model-based planning algorithms Model-based planning: assumes MDP model is available Methods we learned
More informationDynamic Programming: An overview. 1 Preliminaries: The basic principle underlying dynamic programming
Dynamic Programming: An overview These notes summarize some key properties of the Dynamic Programming principle to optimize a function or cost that depends on an interval or stages. This plays a key role
More informationMonte-Carlo Planning: Introduction and Bandit Basics. Alan Fern
Monte-Carlo Planning: Introduction and Bandit Basics Alan Fern 1 Large Worlds We have considered basic model-based planning algorithms Model-based planning: assumes MDP model is available Methods we learned
More informationMarkov Decision Processes
Markov Decision Processes Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. AIMA 3. Chris Amato Stochastic domains So far, we have studied search Can use
More informationMarkov Decision Processes: Making Decision in the Presence of Uncertainty. (some of) R&N R&N
Markov Decision Processes: Making Decision in the Presence of Uncertainty (some of) R&N 16.1-16.6 R&N 17.1-17.4 Different Aspects of Machine Learning Supervised learning Classification - concept learning
More informationHandout 4: Deterministic Systems and the Shortest Path Problem
SEEM 3470: Dynamic Optimization and Applications 2013 14 Second Term Handout 4: Deterministic Systems and the Shortest Path Problem Instructor: Shiqian Ma January 27, 2014 Suggested Reading: Bertsekas
More informationImportance Sampling for Fair Policy Selection
Importance Sampling for Fair Policy Selection Shayan Doroudi Carnegie Mellon University Pittsburgh, PA 15213 shayand@cs.cmu.edu Philip S. Thomas Carnegie Mellon University Pittsburgh, PA 15213 philipt@cs.cmu.edu
More informationDefinition 4.1. In a stochastic process T is called a stopping time if you can tell when it happens.
102 OPTIMAL STOPPING TIME 4. Optimal Stopping Time 4.1. Definitions. On the first day I explained the basic problem using one example in the book. On the second day I explained how the solution to the
More informationEE266 Homework 5 Solutions
EE, Spring 15-1 Professor S. Lall EE Homework 5 Solutions 1. A refined inventory model. In this problem we consider an inventory model that is more refined than the one you ve seen in the lectures. The
More informationReinforcement Learning. Slides based on those used in Berkeley's AI class taught by Dan Klein
Reinforcement Learning Slides based on those used in Berkeley's AI class taught by Dan Klein Reinforcement Learning Basic idea: Receive feedback in the form of rewards Agent s utility is defined by the
More informationIntroduction to Reinforcement Learning. MAL Seminar
Introduction to Reinforcement Learning MAL Seminar 2014-2015 RL Background Learning by interacting with the environment Reward good behavior, punish bad behavior Trial & Error Combines ideas from psychology
More informationReinforcement Learning (1): Discrete MDP, Value Iteration, Policy Iteration
Reinforcement Learning (1): Discrete MDP, Value Iteration, Policy Iteration Piyush Rai CS5350/6350: Machine Learning November 29, 2011 Reinforcement Learning Supervised Learning: Uses explicit supervision
More informationReinforcement Learning (1): Discrete MDP, Value Iteration, Policy Iteration
Reinforcement Learning (1): Discrete MDP, Value Iteration, Policy Iteration Piyush Rai CS5350/6350: Machine Learning November 29, 2011 Reinforcement Learning Supervised Learning: Uses explicit supervision
More informationRollout Allocation Strategies for Classification-based Policy Iteration
Rollout Allocation Strategies for Classification-based Policy Iteration V. Gabillon, A. Lazaric & M. Ghavamzadeh firstname.lastname@inria.fr Workshop on Reinforcement Learning and Search in Very Large
More informationCPSC 540: Machine Learning
CPSC 540: Machine Learning Monte Carlo Methods Mark Schmidt University of British Columbia Winter 2019 Last Time: Markov Chains We can use Markov chains for density estimation, d p(x) = p(x 1 ) p(x }{{}
More informationPhD Qualifier Examination
PhD Qualifier Examination Department of Agricultural Economics May 29, 2015 Instructions This exam consists of six questions. You must answer all questions. If you need an assumption to complete a question,
More informationCHAPTER 5: DYNAMIC PROGRAMMING
CHAPTER 5: DYNAMIC PROGRAMMING Overview This chapter discusses dynamic programming, a method to solve optimization problems that involve a dynamical process. This is in contrast to our previous discussions
More informationCSEP 573: Artificial Intelligence
CSEP 573: Artificial Intelligence Markov Decision Processes (MDP)! Ali Farhadi Many slides over the course adapted from Luke Zettlemoyer, Dan Klein, Pieter Abbeel, Stuart Russell or Andrew Moore 1 Outline
More informationComplex Decisions. Sequential Decision Making
Sequential Decision Making Outline Sequential decision problems Value iteration Policy iteration POMDPs (basic concepts) Slides partially based on the Book "Reinforcement Learning: an introduction" by
More informationCPSC 540: Machine Learning
CPSC 540: Machine Learning Monte Carlo Methods Mark Schmidt University of British Columbia Winter 2018 Last Time: Markov Chains We can use Markov chains for density estimation, p(x) = p(x 1 ) }{{} d p(x
More informationCS 188: Artificial Intelligence Spring Announcements
CS 188: Artificial Intelligence Spring 2011 Lecture 9: MDPs 2/16/2011 Pieter Abbeel UC Berkeley Many slides over the course adapted from either Dan Klein, Stuart Russell or Andrew Moore 1 Announcements
More informationCS 188: Artificial Intelligence
CS 188: Artificial Intelligence Markov Decision Processes Dan Klein, Pieter Abbeel University of California, Berkeley Non-Deterministic Search 1 Example: Grid World A maze-like problem The agent lives
More informationCSE 473: Artificial Intelligence
CSE 473: Artificial Intelligence Markov Decision Processes (MDPs) Luke Zettlemoyer Many slides over the course adapted from Dan Klein, Stuart Russell or Andrew Moore 1 Announcements PS2 online now Due
More informationEE365: Markov Decision Processes
EE365: Markov Decision Processes Markov decision processes Markov decision problem Examples 1 Markov decision processes 2 Markov decision processes add input (or action or control) to Markov chain with
More informationYao s Minimax Principle
Complexity of algorithms The complexity of an algorithm is usually measured with respect to the size of the input, where size may for example refer to the length of a binary word describing the input,
More informationMarkov Decision Processes
Markov Decision Processes Ryan P. Adams COS 324 Elements of Machine Learning Princeton University We now turn to a new aspect of machine learning, in which agents take actions and become active in their
More informationCS 188: Artificial Intelligence
CS 188: Artificial Intelligence Markov Decision Processes Dan Klein, Pieter Abbeel University of California, Berkeley Non Deterministic Search Example: Grid World A maze like problem The agent lives in
More informationProbabilistic Robotics: Probabilistic Planning and MDPs
Probabilistic Robotics: Probabilistic Planning and MDPs Slide credits: Wolfram Burgard, Dieter Fox, Cyrill Stachniss, Giorgio Grisetti, Maren Bennewitz, Christian Plagemann, Dirk Haehnel, Mike Montemerlo,
More informationLecture 2: Making Good Sequences of Decisions Given a Model of World. CS234: RL Emma Brunskill Winter 2018
Lecture 2: Making Good Sequences of Decisions Given a Model of World CS234: RL Emma Brunskill Winter 218 Human in the loop exoskeleton work from Steve Collins lab Class Structure Last Time: Introduction
More informationAdaptive Experiments for Policy Choice. March 8, 2019
Adaptive Experiments for Policy Choice Maximilian Kasy Anja Sautmann March 8, 2019 Introduction The goal of many experiments is to inform policy choices: 1. Job search assistance for refugees: Treatments:
More information4 Reinforcement Learning Basic Algorithms
Learning in Complex Systems Spring 2011 Lecture Notes Nahum Shimkin 4 Reinforcement Learning Basic Algorithms 4.1 Introduction RL methods essentially deal with the solution of (optimal) control problems
More information6.231 DYNAMIC PROGRAMMING LECTURE 3 LECTURE OUTLINE
6.21 DYNAMIC PROGRAMMING LECTURE LECTURE OUTLINE Deterministic finite-state DP problems Backward shortest path algorithm Forward shortest path algorithm Shortest path examples Alternative shortest path
More information91.420/543: Artificial Intelligence UMass Lowell CS Fall 2010
91.420/543: Artificial Intelligence UMass Lowell CS Fall 2010 Lecture 17 & 18: Markov Decision Processes Oct 12 13, 2010 A subset of Lecture 9 slides from Dan Klein UC Berkeley Many slides over the course
More informationCS885 Reinforcement Learning Lecture 3b: May 9, 2018
CS885 Reinforcement Learning Lecture 3b: May 9, 2018 Intro to Reinforcement Learning [SutBar] Sec. 5.1-5.3, 6.1-6.3, 6.5, [Sze] Sec. 3.1, 4.3, [SigBuf] Sec. 2.1-2.5, [RusNor] Sec. 21.1-21.3, CS885 Spring
More informationLecture Notes 1
4.45 Lecture Notes Guido Lorenzoni Fall 2009 A portfolio problem To set the stage, consider a simple nite horizon problem. A risk averse agent can invest in two assets: riskless asset (bond) pays gross
More informationBSc (Hons) Software Engineering BSc (Hons) Computer Science with Network Security
BSc (Hons) Software Engineering BSc (Hons) Computer Science with Network Security Cohorts BCNS/ 06 / Full Time & BSE/ 06 / Full Time Resit Examinations for 2008-2009 / Semester 1 Examinations for 2008-2009
More informationOptimal Stopping. Nick Hay (presentation follows Thomas Ferguson s Optimal Stopping and Applications) November 6, 2008
(presentation follows Thomas Ferguson s and Applications) November 6, 2008 1 / 35 Contents: Introduction Problems Markov Models Monotone Stopping Problems Summary 2 / 35 The Secretary problem You have
More informationMATH3075/3975 FINANCIAL MATHEMATICS TUTORIAL PROBLEMS
MATH307/37 FINANCIAL MATHEMATICS TUTORIAL PROBLEMS School of Mathematics and Statistics Semester, 04 Tutorial problems should be used to test your mathematical skills and understanding of the lecture material.
More informationRegret Minimization against Strategic Buyers
Regret Minimization against Strategic Buyers Mehryar Mohri Courant Institute & Google Research Andrés Muñoz Medina Google Research Motivation Online advertisement: revenue of modern search engine and
More informationLecture 12: MDP1. Victor R. Lesser. CMPSCI 683 Fall 2010
Lecture 12: MDP1 Victor R. Lesser CMPSCI 683 Fall 2010 Biased Random GSAT - WalkSat Notice no random restart 2 Today s lecture Search where there is Uncertainty in Operator Outcome --Sequential Decision
More informationReinforcement Learning and Simulation-Based Search
Reinforcement Learning and Simulation-Based Search David Silver Outline 1 Reinforcement Learning 2 3 Planning Under Uncertainty Reinforcement Learning Markov Decision Process Definition A Markov Decision
More informationReinforcement Learning
Reinforcement Learning Basic idea: Receive feedback in the form of rewards Agent s utility is defined by the reward function Must (learn to) act so as to maximize expected rewards Grid World The agent
More informationUnobserved Heterogeneity Revisited
Unobserved Heterogeneity Revisited Robert A. Miller Dynamic Discrete Choice March 2018 Miller (Dynamic Discrete Choice) cemmap 7 March 2018 1 / 24 Distributional Assumptions about the Unobserved Variables
More informationElif Özge Özdamar T Reinforcement Learning - Theory and Applications February 14, 2006
On the convergence of Q-learning Elif Özge Özdamar elif.ozdamar@helsinki.fi T-61.6020 Reinforcement Learning - Theory and Applications February 14, 2006 the covergence of stochastic iterative algorithms
More informationNon-Deterministic Search
Non-Deterministic Search MDP s 1 Non-Deterministic Search How do you plan (search) when your actions might fail? In general case, how do you plan, when the actions have multiple possible outcomes? 2 Example:
More informationIEOR E4004: Introduction to OR: Deterministic Models
IEOR E4004: Introduction to OR: Deterministic Models 1 Dynamic Programming Following is a summary of the problems we discussed in class. (We do not include the discussion on the container problem or the
More informationPart 4: Markov Decision Processes
Markov decision processes c Vikram Krishnamurthy 2013 1 Part 4: Markov Decision Processes Aim: This part covers discrete time Markov Decision processes whose state is completely observed. The key ideas
More informationCPS 270: Artificial Intelligence Markov decision processes, POMDPs
CPS 270: Artificial Intelligence http://www.cs.duke.edu/courses/fall08/cps270/ Markov decision processes, POMDPs Instructor: Vincent Conitzer Warmup: a Markov process with rewards We derive some reward
More informationLecture 5 January 30
EE 223: Stochastic Estimation and Control Spring 2007 Lecture 5 January 30 Lecturer: Venkat Anantharam Scribe: aryam Kamgarpour 5.1 Secretary Problem The problem set-up is explained in Lecture 4. We review
More information10703 Deep Reinforcement Learning and Control
10703 Deep Reinforcement Learning and Control Russ Salakhutdinov Machine Learning Department rsalakhu@cs.cmu.edu Temporal Difference Learning Used Materials Disclaimer: Much of the material and slides
More informationLecture outline W.B.Powell 1
Lecture outline What is a policy? Policy function approximations (PFAs) Cost function approximations (CFAs) alue function approximations (FAs) Lookahead policies Finding good policies Optimizing continuous
More informationThe Agent-Environment Interface Goals, Rewards, Returns The Markov Property The Markov Decision Process Value Functions Optimal Value Functions
The Agent-Environment Interface Goals, Rewards, Returns The Markov Property The Markov Decision Process Value Functions Optimal Value Functions Optimality and Approximation Finite MDP: {S, A, R, p, γ}
More informationIntro to Reinforcement Learning. Part 3: Core Theory
Intro to Reinforcement Learning Part 3: Core Theory Interactive Example: You are the algorithm! Finite Markov decision processes (finite MDPs) dynamics p p p Experience: S 0 A 0 R 1 S 1 A 1 R 2 S 2 A 2
More informationAM 121: Intro to Optimization Models and Methods
AM 121: Intro to Optimization Models and Methods Lecture 18: Markov Decision Processes Yiling Chen and David Parkes Lesson Plan Markov decision processes Policies and Value functions Solving: average reward,
More informationMonte Carlo Methods (Estimators, On-policy/Off-policy Learning)
1 / 24 Monte Carlo Methods (Estimators, On-policy/Off-policy Learning) Julie Nutini MLRG - Winter Term 2 January 24 th, 2017 2 / 24 Monte Carlo Methods Monte Carlo (MC) methods are learning methods, used
More informationDeep RL and Controls Homework 1 Spring 2017
10-703 Deep RL and Controls Homework 1 Spring 2017 February 1, 2017 Due February 17, 2017 Instructions You have 15 days from the release of the assignment until it is due. Refer to gradescope for the exact
More informationTDT4171 Artificial Intelligence Methods
TDT47 Artificial Intelligence Methods Lecture 7 Making Complex Decisions Norwegian University of Science and Technology Helge Langseth IT-VEST 0 helgel@idi.ntnu.no TDT47 Artificial Intelligence Methods
More informationBasic Framework. About this class. Rewards Over Time. [This lecture adapted from Sutton & Barto and Russell & Norvig]
Basic Framework [This lecture adapted from Sutton & Barto and Russell & Norvig] About this class Markov Decision Processes The Bellman Equation Dynamic Programming for finding value functions and optimal
More informationIntroduction to Dynamic Programming
Introduction to Dynamic Programming http://bicmr.pku.edu.cn/~wenzw/bigdata2018.html Acknowledgement: this slides is based on Prof. Mengdi Wang s and Prof. Dimitri Bertsekas lecture notes Outline 2/65 1
More informationReasoning with Uncertainty
Reasoning with Uncertainty Markov Decision Models Manfred Huber 2015 1 Markov Decision Process Models Markov models represent the behavior of a random process, including its internal state and the externally
More informationMachine Learning in Computer Vision Markov Random Fields Part II
Machine Learning in Computer Vision Markov Random Fields Part II Oren Freifeld Computer Science, Ben-Gurion University March 22, 2018 Mar 22, 2018 1 / 40 1 Some MRF Computations 2 Mar 22, 2018 2 / 40 Few
More informationCS221 / Spring 2018 / Sadigh. Lecture 8: MDPs II
CS221 / Spring 218 / Sadigh Lecture 8: MDPs II cs221.stanford.edu/q Question If you wanted to go from Orbisonia to Rockhill, how would you get there? ride bus 1 ride bus 17 ride the magic tram CS221 /
More informationStochastic Approximation Algorithms and Applications
Harold J. Kushner G. George Yin Stochastic Approximation Algorithms and Applications With 24 Figures Springer Contents Preface and Introduction xiii 1 Introduction: Applications and Issues 1 1.0 Outline
More informationCS 461: Machine Learning Lecture 8
CS 461: Machine Learning Lecture 8 Dr. Kiri Wagstaff kiri.wagstaff@calstatela.edu 2/23/08 CS 461, Winter 2008 1 Plan for Today Review Clustering Reinforcement Learning How different from supervised, unsupervised?
More informationMaking Decisions. CS 3793 Artificial Intelligence Making Decisions 1
Making Decisions CS 3793 Artificial Intelligence Making Decisions 1 Planning under uncertainty should address: The world is nondeterministic. Actions are not certain to succeed. Many events are outside
More informationMaking Complex Decisions
Ch. 17 p.1/29 Making Complex Decisions Chapter 17 Ch. 17 p.2/29 Outline Sequential decision problems Value iteration algorithm Policy iteration algorithm Ch. 17 p.3/29 A simple environment 3 +1 p=0.8 2
More informationCS 188: Artificial Intelligence. Outline
C 188: Artificial Intelligence Markov Decision Processes (MDPs) Pieter Abbeel UC Berkeley ome slides adapted from Dan Klein 1 Outline Markov Decision Processes (MDPs) Formalism Value iteration In essence
More informationAdmissioncontrolwithbatcharrivals
Admissioncontrolwithbatcharrivals E. Lerzan Örmeci Department of Industrial Engineering Koç University Sarıyer 34450 İstanbul-Turkey Apostolos Burnetas Department of Operations Weatherhead School of Management
More informationQ1. [?? pts] Search Traces
CS 188 Spring 2010 Introduction to Artificial Intelligence Midterm Exam Solutions Q1. [?? pts] Search Traces Each of the trees (G1 through G5) was generated by searching the graph (below, left) with a
More information2D5362 Machine Learning
2D5362 Machine Learning Reinforcement Learning MIT GALib Available at http://lancet.mit.edu/ga/ download galib245.tar.gz gunzip galib245.tar.gz tar xvf galib245.tar cd galib245 make or access my files
More informationIntroduction to Fall 2007 Artificial Intelligence Final Exam
NAME: SID#: Login: Sec: 1 CS 188 Introduction to Fall 2007 Artificial Intelligence Final Exam You have 180 minutes. The exam is closed book, closed notes except a two-page crib sheet, basic calculators
More informationReinforcement Learning
Reinforcement Learning Monte Carlo Methods Heiko Zimmermann 15.05.2017 1 Monte Carlo Monte Carlo policy evaluation First visit policy evaluation Estimating q values On policy methods Off policy methods
More informationLogistics. CS 473: Artificial Intelligence. Markov Decision Processes. PS 2 due today Midterm in one week
CS 473: Artificial Intelligence Markov Decision Processes Dan Weld University of Washington [Slides originally created by Dan Klein & Pieter Abbeel for CS188 Intro to AI at UC Berkeley. All CS188 materials
More informationMonte-Carlo Planning: Basic Principles and Recent Progress
Monte-Carlo Planning: Basic Principles and Recent Progress Alan Fern School of EECS Oregon State University Outline Preliminaries: Markov Decision Processes What is Monte-Carlo Planning? Uniform Monte-Carlo
More informationMaximum Contiguous Subsequences
Chapter 8 Maximum Contiguous Subsequences In this chapter, we consider a well-know problem and apply the algorithm-design techniques that we have learned thus far to this problem. While applying these
More informationDynamic Programming (DP) Massimo Paolucci University of Genova
Dynamic Programming (DP) Massimo Paolucci University of Genova DP cannot be applied to each kind of problem In particular, it is a solution method for problems defined over stages For each stage a subproblem
More informationFinal exam solutions
EE365 Stochastic Control / MS&E251 Stochastic Decision Models Profs. S. Lall, S. Boyd June 5 6 or June 6 7, 2013 Final exam solutions This is a 24 hour take-home final. Please turn it in to one of the
More informationMartingale Pricing Theory in Discrete-Time and Discrete-Space Models
IEOR E4707: Foundations of Financial Engineering c 206 by Martin Haugh Martingale Pricing Theory in Discrete-Time and Discrete-Space Models These notes develop the theory of martingale pricing in a discrete-time,
More informationDecision Theory: Value Iteration
Decision Theory: Value Iteration CPSC 322 Decision Theory 4 Textbook 9.5 Decision Theory: Value Iteration CPSC 322 Decision Theory 4, Slide 1 Lecture Overview 1 Recap 2 Policies 3 Value Iteration Decision
More informationCE 191: Civil and Environmental Engineering Systems Analysis. LEC 15 : DP Examples
CE 191: Civil and Environmental Engineering Systems Analysis LEC 15 : DP Examples Professor Scott Moura Civil & Environmental Engineering University of California, Berkeley Fall 2014 Prof. Moura UC Berkeley
More informationOptimization Methods. Lecture 16: Dynamic Programming
15.093 Optimization Methods Lecture 16: Dynamic Programming 1 Outline 1. The knapsack problem Slide 1. The traveling salesman problem 3. The general DP framework 4. Bellman equation 5. Optimal inventory
More informationComputer Vision Group Prof. Daniel Cremers. 7. Sequential Data
Group Prof. Daniel Cremers 7. Sequential Data Bayes Filter (Rep.) We can describe the overall process using a Dynamic Bayes Network: This incorporates the following Markov assumptions: (measurement) (state)!2
More informationIntelligent Systems (AI-2)
Intelligent Systems (AI-2) Computer Science cpsc422, Lecture 9 Sep, 28, 2016 Slide 1 CPSC 422, Lecture 9 An MDP Approach to Multi-Category Patient Scheduling in a Diagnostic Facility Adapted from: Matthew
More informationAlgorithmic Trading using Reinforcement Learning augmented with Hidden Markov Model
Algorithmic Trading using Reinforcement Learning augmented with Hidden Markov Model Simerjot Kaur (sk3391) Stanford University Abstract This work presents a novel algorithmic trading system based on reinforcement
More informationDynamic Portfolio Choice II
Dynamic Portfolio Choice II Dynamic Programming Leonid Kogan MIT, Sloan 15.450, Fall 2010 c Leonid Kogan ( MIT, Sloan ) Dynamic Portfolio Choice II 15.450, Fall 2010 1 / 35 Outline 1 Introduction to Dynamic
More informationReinforcement Learning and Optimal Control. Chapter 1 Exact Dynamic Programming DRAFT
Reinforcement Learning and Optimal Control by Dimitri P. Bertsekas Massachusetts Institute of Technology Chapter 1 Exact Dynamic Programming DRAFT This is Chapter 1 of the draft textbook Reinforcement
More informationMonte-Carlo Planning Look Ahead Trees. Alan Fern
Monte-Carlo Planning Look Ahead Trees Alan Fern 1 Monte-Carlo Planning Outline Single State Case (multi-armed bandits) A basic tool for other algorithms Monte-Carlo Policy Improvement Policy rollout Policy
More informationScenario Generation and Sampling Methods
Scenario Generation and Sampling Methods Güzin Bayraksan Tito Homem-de-Mello SVAN 2016 IMPA May 9th, 2016 Bayraksan (OSU) & Homem-de-Mello (UAI) Scenario Generation and Sampling SVAN IMPA May 9 1 / 30
More informationSOLVING ROBUST SUPPLY CHAIN PROBLEMS
SOLVING ROBUST SUPPLY CHAIN PROBLEMS Daniel Bienstock Nuri Sercan Özbay Columbia University, New York November 13, 2005 Project with Lucent Technologies Optimize the inventory buffer levels in a complicated
More informationarxiv: v1 [math.pr] 6 Apr 2015
Analysis of the Optimal Resource Allocation for a Tandem Queueing System arxiv:1504.01248v1 [math.pr] 6 Apr 2015 Liu Zaiming, Chen Gang, Wu Jinbiao School of Mathematics and Statistics, Central South University,
More informationCS221 / Autumn 2018 / Liang. Lecture 8: MDPs II
CS221 / Autumn 218 / Liang Lecture 8: MDPs II cs221.stanford.edu/q Question If you wanted to go from Orbisonia to Rockhill, how would you get there? ride bus 1 ride bus 17 ride the magic tram CS221 / Autumn
More informationMonte-Carlo Planning Look Ahead Trees. Alan Fern
Monte-Carlo Planning Look Ahead Trees Alan Fern 1 Monte-Carlo Planning Outline Single State Case (multi-armed bandits) A basic tool for other algorithms Monte-Carlo Policy Improvement Policy rollout Policy
More informationCS221 / Spring 2018 / Sadigh. Lecture 9: Games I
CS221 / Spring 2018 / Sadigh Lecture 9: Games I Course plan Search problems Markov decision processes Adversarial games Constraint satisfaction problems Bayesian networks Reflex States Variables Logic
More information