Transfer Learning and Intelligence: an Argument and Approach Matthew E. Taylor Joint work with: Gregory Kuhlmann and Peter Stone Learning Agents Research.

Slides:



Advertisements
Similar presentations
Heuristic Search techniques
Advertisements

Meta-Level Control in Multi-Agent Systems Anita Raja and Victor Lesser Department of Computer Science University of Massachusetts Amherst, MA
Extraction and Transfer of Knowledge in Reinforcement Learning A.LAZARIC Inria “30 minutes de Science” Seminars SequeL Inria Lille – Nord Europe December.
Ai in game programming it university of copenhagen Reinforcement Learning [Outro] Marco Loog.
Background Reinforcement Learning (RL) agents learn to do tasks by iteratively performing actions in the world and using resulting experiences to decide.
01 -1 Lecture 01 Artificial Intelligence Topics –Introduction –Knowledge representation –Knowledge reasoning –Machine learning –Applications.
An Introduction to Machine Learning In the area of AI (earlier) machine learning took a back seat to Expert Systems Expert system development usually consists.
Algorithmic Problems in Algebraic Structures Undecidability Paul Bell Supervisor: Dr. Igor Potapov Department of Computer Science
Reinforcement Learning (1)
Reinforcement Learning of Local Shape in the Game of Atari-Go David Silver.
Intelligent Agents: an Overview. 2 Definitions Rational behavior: to achieve a goal minimizing the cost and maximizing the satisfaction. Rational agent:
Marco Adelfio CMSC 828N – Spring 2009 General Game Playing (GGP)
CS Machine Learning. What is Machine Learning? Adapt to / learn from data  To optimize a performance function Can be used to:  Extract knowledge.
Reinforcement Learning in the Presence of Hidden States Andrew Howard Andrew Arnold {ah679
Lisa Torrey University of Wisconsin – Madison CS 540.
Chapter 14: Artificial Intelligence Invitation to Computer Science, C++ Version, Third Edition.
ARTIFICIAL INTELLIGENCE [INTELLIGENT AGENTS PARADIGM] Professor Janis Grundspenkis Riga Technical University Faculty of Computer Science and Information.
Artificial Intelligence Introductory Lecture Jennifer J. Burg Department of Mathematics and Computer Science.
Skill Acquisition via Transfer Learning and Advice Taking Lisa Torrey, Jude Shavlik, Trevor Walker University of Wisconsin-Madison, USA Richard Maclin.
1 ECE-517 Reinforcement Learning in Artificial Intelligence Lecture 7: Finite Horizon MDPs, Dynamic Programming Dr. Itamar Arel College of Engineering.
Lecture 10: 8/6/1435 Machine Learning Lecturer/ Kawther Abas 363CS – Artificial Intelligence.
Leveraging Human Knowledge for Machine Learning Curriculum Design Matthew E. Taylor teamcore.usc.edu/taylorm.
Reinforcement Learning 主講人:虞台文 Content Introduction Main Elements Markov Decision Process (MDP) Value Functions.
Design Principles for Creating Human-Shapable Agents W. Bradley Knox, Ian Fasel, and Peter Stone The University of Texas at Austin Department of Computer.
Curiosity-Driven Exploration with Planning Trajectories Tyler Streeter PhD Student, Human Computer Interaction Iowa State University
POMDPs: 5 Reward Shaping: 4 Intrinsic RL: 4 Function Approximation: 3.
Kansas State University Department of Computing and Information Sciences CIS 730: Introduction to Artificial Intelligence Lecture 9 of 42 Wednesday, 14.
Rational Agency CSMC Introduction to Artificial Intelligence January 8, 2004.
Copyright Paula Matuszek Kinds of Machine Learning.
Transfer Learning in Sequential Decision Problems: A Hierarchical Bayesian Approach Aaron Wilson, Alan Fern, Prasad Tadepalli School of EECS Oregon State.
The ideals reality of science The pursuit of verifiable answers highly cited papers for your c.v. The validation of our results by reproduction convincing.
Learning for Physically Diverse Robot Teams Robot Teams - Chapter 7 CS8803 Autonomous Multi-Robot Systems 10/3/02.
Flowcharts C++ Lab. Algorithm An informal definition of an algorithm is: a step-by-step method for solving a problem or doing a task. Input data A step-by-step.
Any Questions? Programming Assignments?. CptS 450/516 Singularity Weka / Torch / RL-Glue/Burlap/RLPy Finite MDP = ? Start arbitrarily, moving towards.
Matthew E. Taylor 1 Autonomous Inter-Task Transfer in Reinforcement Learning Domains Matthew E. Taylor Learning Agents Research Group Department of Computer.
Reinforcement Learning Guest Lecturer: Chengxiang Zhai Machine Learning December 6, 2001.
Algorithmic Problems in Algebraic Structures Undecidability Paul Bell Supervisor: Dr. Igor Potapov Department of Computer Science
REINFORCEMENT LEARNING Unsupervised learning 1. 2 So far ….  Supervised machine learning: given a set of annotated istances and a set of categories,
CS 5751 Machine Learning Chapter 13 Reinforcement Learning1 Reinforcement Learning Control learning Control polices that choose optimal actions Q learning.
Brief Intro to Machine Learning CS539
Chapter 11: Artificial Intelligence
Deep Reinforcement Learning
A Comparison of Learning Algorithms on the ALE
Eick: Introduction Machine Learning
Spring 2003 Dr. Susan Bridges
Done Done Course Overview What is AI? What are the Major Challenges?
Learning to Generate Networks
Presented By S.Yamuna AP/CSE
Reinforcement Learning (1)
Reinforcement learning (Chapter 21)
Transferring Instances for Model-Based Reinforcement Learning
Intelligent Agents Chapter 2.
© James D. Skrentny from notes by C. Dyer, et. al.
Louise Lawrence Supervisor: Dr Wolfgang Mayer
Convergence, Targeted Optimality, and Safety in Multiagent Learning
CS 188: Artificial Intelligence
Reinforcement Learning
Announcements Homework 3 due today (grace period through Friday)
CSE 4705 Artificial Intelligence
CAP 5636 – Advanced Artificial Intelligence
Dr. Unnikrishnan P.C. Professor, EEE
October 6, 2011 Dr. Itamar Arel College of Engineering
Introduction to Reinforcement Learning and Q-Learning
Emir Zeylan Stylianos Filippou
Christoph F. Eick: A Gentle Introduction to Machine Learning
Introduction to Artificial Intelligence Instructor: Dr. Eduardo Urbina
Modeling and Analysis Tutorial
Markov Decision Processes
Angel A. Cantu, Nami Akazawa Department of Computer Science
Markov Decision Processes
Presentation transcript:

Transfer Learning and Intelligence: an Argument and Approach Matthew E. Taylor Joint work with: Gregory Kuhlmann and Peter Stone Learning Agents Research Group Department of Computer Sciences The University of Texas at Austin

Result Summary: AGI-08 Help select source task for given target Transfer a search heuristic Target Task Size Hand-coded Heuristic? Target Task Random Start State? Target Task Random Goal State? Transfer Percentage 8x8YesNo 75 8x8No 73 10x10No 66 8x8NoYesNo69 8x8NoYesYes (anywhere)92 8x8NoYesYes (SW quadrant)70

AGI & Learning Why Learn? 1.Better solutions 2.On-line adaptation Current Problems: 1.Commonly applied to simple tasks 2.Algorithms often data-inefficient 3.Need substantial amounts of human knowledge One possible answer: Transfer Learning

Transfer Learning (related to Lifelong Learning or Multi-task Learning) Learn across multiple tasks: Learn faster Harder tasks become tractable Learn with less human input Prerequisite for AGI?

Transfer Examples Learn difficult tasks faster – Learn a set of simple tasks – Eventually learn target task – Total time reduction Autonomous transfer – Explore the world, learning – Transfer autonomously – Effectively use past knowledge

Transfer in Reinforcement Learning Environment Agent Action State Reward Source Task Environment Agent Action StateReward Target Task

Representative Transfer Results

Policy: π(s) → a Action-value function: Q(s,a) → R Model of the environment: T(s, a) → s’ Rules / Advice Higher-level information – Search heuristic – Learning rates – Appropriate features What to transfer? Environment Agent Action StateReward

How to transfer? Human design (engineering task) Construct a sequence of tasks Provide learner with mappings between tasks Fully autonomous (not yet achieved) Learn if tasks are related Learn how tasks are related ? ?

Result Summary: AGI-08 Help select source task for given target Transfer a search heuristic General Game Playing task W13: Transfer Learning for Complex Tasks