Natural Actor-Critic. Authors: Jan Peters and Stefan Schaal Neurocomputing, Cognitive robotics 2008/2009 Wouter Klijn

Size: px
Start display at page:

Download "Natural Actor-Critic. Authors: Jan Peters and Stefan Schaal Neurocomputing, Cognitive robotics 2008/2009 Wouter Klijn"

Transcription

1 Natural Actor-Critic Authors: Jan Peters and Stefan Schaal Neurocomputing, 2008 Cognitive robotics 2008/2009 Wouter Klijn

2 Content Content / Introduction Actor-Critic Natural gradient Applications Conclusion References

3 Actor-Critic Separate memory structure for policy (Actor) and value function (Critic). After each action the critic evaluates the new state and returns an error. The actor and the critic are updated using this error. The Actor-Critic Architecture [2]

4 Actor-Critic: Notation The model in the article is based loosely on a MDP. - Discrete time - Continues state set: - Continues action set: The system: - Start state: drawn from a start-state distribution - At any state the actor chooses an action -The system transfers to a new state. -The system yields a reward after each action.

5 Actor-Critic: Functions Goal of the system is to find a policy This goal is reached by optimizing the normalized expected return as a function of the inputs With the differential : Problem: The meat and bones of the article gets lost in convoluted functions. Solution: Use a (presumably) known model/system that can be improved using the same method [4].

6 Actor-Critic: Simplified model Actor: - Universal function approximator e.g. Multi Layer Perceptron (MLP). - Gets error from the critic. - Gradient descent! Critic: Baseline (based on example data or a constant) times a function containing learned information combined with the reward.

7 Natural Gradient : Vanilla Gradient Descent The critic returns an error which in combination with the function approximator can be used to create an error function The partial differential of this error function, the gradient can now be used to update the internal variables in the function approximator (and critic). Gradient descent [3]

8 Natural Gradient: definition An alternative gradient to update the function approximator. Definition of the natural gradient: Where Matrix (FIM). denotes the transposed Fisher Information The FIM is a statistical construct that summarizes the mean and variation of the input data. Used in combination with the natural gradient FIM gives the direction of steepest descent [4].

9 Natural Gradient: Properties The natural gradient is a linear weighted version of the normal (vanilla) gradient. Convergence to a local minimum is guaranteed. By choosing a more direct path to optimal solution faster convergence is reached avoiding premature convergence. Covariant: Independent of the coordinate frame. Averages out stochasticity resulting in smaller datasets for estimating the correct data set. Gradient landscape for the vanilla and natural gradient. Adapted from [1]

10 Natural Gradient: plateaus The natural gradient is a solution to escape from plateaus in the gradient landscape. Plateaus are parts where the gradients of a function are extremely small. It takes considerate time to traverse these and are well know feature of gradient descent methods. Example function landscape showing multiple plateaus and the resulting error while traversing it with normal gradient steps (iterations) [5]

11 Applications: Cart-Pole Balancing Well known benchmark for reinforced learning [1] Unstable non-linear system that can be simulated. State: Action: Reward based on the current state with constant baseline. (Episodic Actor-Critic)

12 Applications: Cart-Pole Balancing Simulated experiment with a sample rate of 60 hertz, Comparing natural and vanilla gradient Actor-Critic algorithms. Results: The natural gradient implementation takes on average ten minutes to find an optimal solution. The vanilla gradient takes on average two hours. Expected return policy error averaged over 100 simulated runs

13 Applications: Baseball Optimizing nonlinear dynamic motor primitives for robotics. In plain English: Learning a robot to hit a ball. Shows the usage of a baseline for the critic: A teacher manipulating the robot. (LSTD-Q(λ) Actor-Critic) State, action and reward not explicitly given but are based on the motor primitives (and presumably a camera input): Optimal (red), POMDP (dashed) and Actor-Critic motor primitives.

14 Applications: Baseball The task of the robot is to hit the ball so that is flies as far as possible. The robot has seven degrees of freedom. Initially the robot is taught by supervised learning and fails. Subsequently the performance is improved by the Natural Actor-critic.

15 Applications: Baseball Both learning methods eventually learn their version of the best solution. However the POMDB requires 10^6 learning steps compared to 10^3 for the Natural Actor-Critic. Remarkable is that the Natural Actor-critic subjectively has a solution that is closer to the teacher/optimal solution.

16 Conclusions A novel policy-gradient reinforcement learning method. Two distinct flavors: -Episodic with a constant as a baseline function in the critic - LSTD-Q(λ) with a rich baseline (teacher) function. The improved functioning can be traced back to the usage of the improved natural gradient which uses statistical information of the input data to optimize changes in the used learning functions.

17 Conclusions The preliminary versions of the method have been implemented in wide range of real word applications: - Humanoid robots -Trafic light optimalisation - Multirobot systems - Gait optimalisation in robot locomotion.

18 References [1] J. Peters and S. Schaal, Natural Actor Critic. Neurocomputing, 2008 [2] R.S. Sutton and A.G. Barto, Reinforcement Learning: An Introduction MIT Press, Cambridge, 1998 Web version: [3] [4] S. Amari Natural Gradient Works Efficiently in Learning Neural Computation 10, (1998) [5] K. Fukumizu, S. Amari Local Minima and Plateaus in Hierarchical Structures of Multilayer Perceptrons, Neural Networks, 2000

Applying the Episodic Natural Actor-Critic Architecture to Motor Primitive Learning

Applying the Episodic Natural Actor-Critic Architecture to Motor Primitive Learning Applying the Episodic Natural Actor-Critic Architecture to Motor Primitive Learning Jan Peters 1, Stefan Schaal 1 University of Southern California, Los Angeles CA 90089, USA Abstract. In this paper, we

More information

10703 Deep Reinforcement Learning and Control

10703 Deep Reinforcement Learning and Control 10703 Deep Reinforcement Learning and Control Russ Salakhutdinov Machine Learning Department rsalakhu@cs.cmu.edu Policy Gradient I Used Materials Disclaimer: Much of the material and slides for this lecture

More information

Learning Motor Behaviors: Past & Present Work

Learning Motor Behaviors: Past & Present Work Stefan Schaal Computer Science & Neuroscience University of Southern California, Los Angeles & ATR Computational Neuroscience Laboratory Kyoto, Japan sschaal@usc.edu http://www-clmc.usc.edu Learning Motor

More information

Deep Reinforcement Learning

Deep Reinforcement Learning Deep Reinforcement Learning 1 Outline 1. Overview of Reinforcement Learning 2. Policy Search 3. Policy Gradient and Gradient Estimators 4. Q-prop: Sample Efficient Policy Gradient and an Off-policy Critic

More information

Introduction to Reinforcement Learning. J. Zico Kolter Carnegie Mellon University

Introduction to Reinforcement Learning. J. Zico Kolter Carnegie Mellon University Introduction to Reinforcement Learning J. Zico Kolter Carnegie Mellon University 1 Agent interaction with environment Agent State s Reward r Action a Environment 2 Of course, an oversimplification 3 Review:

More information

10703 Deep Reinforcement Learning and Control

10703 Deep Reinforcement Learning and Control 10703 Deep Reinforcement Learning and Control Russ Salakhutdinov Machine Learning Department rsalakhu@cs.cmu.edu Policy Gradient II Used Materials Disclaimer: Much of the material and slides for this lecture

More information

Learning to bounce a ball with a robotic arm

Learning to bounce a ball with a robotic arm Eric Wolter TU Darmstadt Thorsten Baark TU Darmstadt Abstract Bouncing a ball is a fun and challenging task for humans. It requires fine and complex motor controls and thus is an interesting problem for

More information

Two steps Natural Actor Critic Learning for Underwater Cable Tracking

Two steps Natural Actor Critic Learning for Underwater Cable Tracking 2 IEEE International Conference on Robotics and Automation Anchorage Convention District May 3-8, 2, Anchorage, Alaska, USA Two steps Natural Actor Critic Learning for Underwater Cable Tracking Andres

More information

A Brief Introduction to Reinforcement Learning

A Brief Introduction to Reinforcement Learning A Brief Introduction to Reinforcement Learning Minlie Huang ( ) Dept. of Computer Science, Tsinghua University aihuang@tsinghua.edu.cn 1 http://coai.cs.tsinghua.edu.cn/hml Reinforcement Learning Agent

More information

Approximating a Policy Can be Easier Than Approximating a Value Function

Approximating a Policy Can be Easier Than Approximating a Value Function Computer Science Technical Report Approximating a Policy Can be Easier Than Approximating a Value Function Charles W. Anderson www.cs.colo.edu/ anderson February, 2 Technical Report CS-- Computer Science

More information

Reinforcement Learning and Shape Grammars

Reinforcement Learning and Shape Grammars Reinforcement Learning and Shape Grammars Technical report Author Manuela Ruiz Montiel Date April 15, 2011 Version 1.0 1 Contents 0. Introduction... 3 1. Tabular approach... 4 1.1 Tabular Q-learning...

More information

MEMORY AUGMENTED CONTROL NETWORKS

MEMORY AUGMENTED CONTROL NETWORKS MEMORY AUGMENTED CONTROL NETWORKS Arbaaz Khan, Clark Zhang, Nikolay Atanasov, Konstantinos Karydis, Vijay Kumar, Daniel D. Lee GRASP Laboratory, University of Pennsylvania Presented by Aravind Balakrishnan

More information

Reinforcement Learning for Appearance Based Visual Servoing in Robotic Manipulation

Reinforcement Learning for Appearance Based Visual Servoing in Robotic Manipulation Reinforcement Learning for Appearance Based Visual Servoing in Robotic Manipulation UMAR KHAN, LIAQUAT ALI KHAN, S. ZAHID HUSSAIN Department of Mechatronics Engineering AIR University E-9, Islamabad PAKISTAN

More information

Learning. Learning agents Inductive learning. Neural Networks. Different Learning Scenarios Evaluation

Learning. Learning agents Inductive learning. Neural Networks. Different Learning Scenarios Evaluation Learning Learning agents Inductive learning Different Learning Scenarios Evaluation Slides based on Slides by Russell/Norvig, Ronald Williams, and Torsten Reil Material from Russell & Norvig, chapters

More information

Reinforcement Learning (INF11010) Lecture 6: Dynamic Programming for Reinforcement Learning (extended)

Reinforcement Learning (INF11010) Lecture 6: Dynamic Programming for Reinforcement Learning (extended) Reinforcement Learning (INF11010) Lecture 6: Dynamic Programming for Reinforcement Learning (extended) Pavlos Andreadis, February 2 nd 2018 1 Markov Decision Processes A finite Markov Decision Process

More information

Unsupervised Learning: Clustering

Unsupervised Learning: Clustering Unsupervised Learning: Clustering Vibhav Gogate The University of Texas at Dallas Slides adapted from Carlos Guestrin, Dan Klein & Luke Zettlemoyer Machine Learning Supervised Learning Unsupervised Learning

More information

Learn to Swing Up and Balance a Real Pole Based on Raw Visual Input Data

Learn to Swing Up and Balance a Real Pole Based on Raw Visual Input Data Learn to Swing Up and Balance a Real Pole Based on Raw Visual Input Data Jan Mattner*, Sascha Lange, and Martin Riedmiller Machine Learning Lab Department of Computer Science University of Freiburg 79110,

More information

Actor-Critic Control with Reference Model Learning

Actor-Critic Control with Reference Model Learning Actor-Critic Control with Reference Model Learning Ivo Grondman Maarten Vaandrager Lucian Buşoniu Robert Babuška Erik Schuitema Delft Center for Systems and Control, Faculty 3mE, Delft University of Technology,

More information

arxiv: v1 [cs.ro] 10 May 2014

arxiv: v1 [cs.ro] 10 May 2014 Efficient Reuse of Previous Experiences to Improve Policies in Real Environment arxiv:1405.2406v1 [cs.ro] 10 May 2014 Norikazu Sugimoto 1,3, Voot Tangkaratt 2, Thijs Wensveen 4, Tingting Zhao 2, Masashi

More information

Reinforcement Learning for Parameterized Motor Primitives

Reinforcement Learning for Parameterized Motor Primitives 2006 International Joint Conference on Neural Networks Sheraton Vancouver Wall Centre Hotel, Vancouver, BC, Canada July 16-21, 2006 Reinforcement Learning for Parameterized Motor Primitives Jan Peters

More information

PATTERN CLASSIFICATION AND SCENE ANALYSIS

PATTERN CLASSIFICATION AND SCENE ANALYSIS PATTERN CLASSIFICATION AND SCENE ANALYSIS RICHARD O. DUDA PETER E. HART Stanford Research Institute, Menlo Park, California A WILEY-INTERSCIENCE PUBLICATION JOHN WILEY & SONS New York Chichester Brisbane

More information

Kinesthetic Teaching via Fast Marching Square

Kinesthetic Teaching via Fast Marching Square Kinesthetic Teaching via Fast Marching Square Javier V. Gómez, David Álvarez, Santiago Garrido and Luis Moreno Abstract This paper presents a novel robotic learning technique based on Fast Marching Square

More information

Using Continuous Action Spaces to Solve Discrete Problems

Using Continuous Action Spaces to Solve Discrete Problems Using Continuous Action Spaces to Solve Discrete Problems Hado van Hasselt Marco A. Wiering Abstract Real-world control problems are often modeled as Markov Decision Processes (MDPs) with discrete action

More information

ADAPTIVE TILE CODING METHODS FOR THE GENERALIZATION OF VALUE FUNCTIONS IN THE RL STATE SPACE A THESIS SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL

ADAPTIVE TILE CODING METHODS FOR THE GENERALIZATION OF VALUE FUNCTIONS IN THE RL STATE SPACE A THESIS SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL ADAPTIVE TILE CODING METHODS FOR THE GENERALIZATION OF VALUE FUNCTIONS IN THE RL STATE SPACE A THESIS SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA BY BHARAT SIGINAM IN

More information

The Organization of Cortex-Ganglia-Thalamus to Generate Movements From Motor Primitives: a Model for Developmental Robotics

The Organization of Cortex-Ganglia-Thalamus to Generate Movements From Motor Primitives: a Model for Developmental Robotics The Organization of Cortex-Ganglia-Thalamus to Generate Movements From Motor Primitives: a Model for Developmental Robotics Alessio Mauro Franchi 1, Danilo Attuario 2, and Giuseppina Gini 3 Department

More information

Lecture 1 Notes. Outline. Machine Learning. What is it? Instructors: Parth Shah, Riju Pahwa

Lecture 1 Notes. Outline. Machine Learning. What is it? Instructors: Parth Shah, Riju Pahwa Instructors: Parth Shah, Riju Pahwa Lecture 1 Notes Outline 1. Machine Learning What is it? Classification vs. Regression Error Training Error vs. Test Error 2. Linear Classifiers Goals and Motivations

More information

Human-level Control Through Deep Reinforcement Learning (Deep Q Network) Peidong Wang 11/13/2015

Human-level Control Through Deep Reinforcement Learning (Deep Q Network) Peidong Wang 11/13/2015 Human-level Control Through Deep Reinforcement Learning (Deep Q Network) Peidong Wang 11/13/2015 Content Demo Framework Remarks Experiment Discussion Content Demo Framework Remarks Experiment Discussion

More information

WestminsterResearch

WestminsterResearch WestminsterResearch http://www.westminster.ac.uk/research/westminsterresearch Reinforcement learning in continuous state- and action-space Barry D. Nichols Faculty of Science and Technology This is an

More information

Topics in AI (CPSC 532L): Multimodal Learning with Vision, Language and Sound. Lecture 12: Deep Reinforcement Learning

Topics in AI (CPSC 532L): Multimodal Learning with Vision, Language and Sound. Lecture 12: Deep Reinforcement Learning Topics in AI (CPSC 532L): Multimodal Learning with Vision, Language and Sound Lecture 12: Deep Reinforcement Learning Types of Learning Supervised training Learning from the teacher Training data includes

More information

A Systematic Overview of Data Mining Algorithms

A Systematic Overview of Data Mining Algorithms A Systematic Overview of Data Mining Algorithms 1 Data Mining Algorithm A well-defined procedure that takes data as input and produces output as models or patterns well-defined: precisely encoded as a

More information

Neural Network Learning. Today s Lecture. Continuation of Neural Networks. Artificial Neural Networks. Lecture 24: Learning 3. Victor R.

Neural Network Learning. Today s Lecture. Continuation of Neural Networks. Artificial Neural Networks. Lecture 24: Learning 3. Victor R. Lecture 24: Learning 3 Victor R. Lesser CMPSCI 683 Fall 2010 Today s Lecture Continuation of Neural Networks Artificial Neural Networks Compose of nodes/units connected by links Each link has a numeric

More information

Residual Advantage Learning Applied to a Differential Game

Residual Advantage Learning Applied to a Differential Game Presented at the International Conference on Neural Networks (ICNN 96), Washington DC, 2-6 June 1996. Residual Advantage Learning Applied to a Differential Game Mance E. Harmon Wright Laboratory WL/AAAT

More information

Teaching a robot to perform a basketball shot using EM-based reinforcement learning methods

Teaching a robot to perform a basketball shot using EM-based reinforcement learning methods Teaching a robot to perform a basketball shot using EM-based reinforcement learning methods Tobias Michels TU Darmstadt Aaron Hochländer TU Darmstadt Abstract In this paper we experiment with reinforcement

More information

Neuro-Dynamic Programming An Overview

Neuro-Dynamic Programming An Overview 1 Neuro-Dynamic Programming An Overview Dimitri Bertsekas Dept. of Electrical Engineering and Computer Science M.I.T. May 2006 2 BELLMAN AND THE DUAL CURSES Dynamic Programming (DP) is very broadly applicable,

More information

MODIFIED KALMAN FILTER BASED METHOD FOR TRAINING STATE-RECURRENT MULTILAYER PERCEPTRONS

MODIFIED KALMAN FILTER BASED METHOD FOR TRAINING STATE-RECURRENT MULTILAYER PERCEPTRONS MODIFIED KALMAN FILTER BASED METHOD FOR TRAINING STATE-RECURRENT MULTILAYER PERCEPTRONS Deniz Erdogmus, Justin C. Sanchez 2, Jose C. Principe Computational NeuroEngineering Laboratory, Electrical & Computer

More information

Locally Weighted Learning for Control. Alexander Skoglund Machine Learning Course AASS, June 2005

Locally Weighted Learning for Control. Alexander Skoglund Machine Learning Course AASS, June 2005 Locally Weighted Learning for Control Alexander Skoglund Machine Learning Course AASS, June 2005 Outline Locally Weighted Learning, Christopher G. Atkeson et. al. in Artificial Intelligence Review, 11:11-73,1997

More information

Approximate Linear Successor Representation

Approximate Linear Successor Representation Approximate Linear Successor Representation Clement A. Gehring Computer Science and Artificial Intelligence Laboratory Massachusetts Institutes of Technology Cambridge, MA 2139 gehring@csail.mit.edu Abstract

More information

Neuro-Fuzzy Inverse Forward Models

Neuro-Fuzzy Inverse Forward Models CS9 Autumn Neuro-Fuzzy Inverse Forward Models Brian Highfill Stanford University Department of Computer Science Abstract- Internal cognitive models are useful methods for the implementation of motor control

More information

ICA as a preprocessing technique for classification

ICA as a preprocessing technique for classification ICA as a preprocessing technique for classification V.Sanchez-Poblador 1, E. Monte-Moreno 1, J. Solé-Casals 2 1 TALP Research Center Universitat Politècnica de Catalunya (Catalonia, Spain) enric@gps.tsc.upc.es

More information

A Systematic Overview of Data Mining Algorithms. Sargur Srihari University at Buffalo The State University of New York

A Systematic Overview of Data Mining Algorithms. Sargur Srihari University at Buffalo The State University of New York A Systematic Overview of Data Mining Algorithms Sargur Srihari University at Buffalo The State University of New York 1 Topics Data Mining Algorithm Definition Example of CART Classification Iris, Wine

More information

Reinforcement Learning with Parameterized Actions

Reinforcement Learning with Parameterized Actions Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (AAAI-16) Reinforcement Learning with Parameterized Actions Warwick Masson and Pravesh Ranchod School of Computer Science and Applied

More information

291 Programming Assignment #3

291 Programming Assignment #3 000 001 002 003 004 005 006 007 008 009 010 011 012 013 014 015 016 017 018 019 020 021 022 023 024 025 026 027 028 029 030 031 032 033 034 035 036 037 038 039 040 041 042 043 044 045 046 047 048 049 050

More information

Bilevel Sparse Coding

Bilevel Sparse Coding Adobe Research 345 Park Ave, San Jose, CA Mar 15, 2013 Outline 1 2 The learning model The learning algorithm 3 4 Sparse Modeling Many types of sensory data, e.g., images and audio, are in high-dimensional

More information

Getting a kick out of humanoid robotics

Getting a kick out of humanoid robotics Getting a kick out of humanoid robotics Using reinforcement learning to shape a soccer kick Christiaan W. Meijer Getting a kick out of humanoid robotics Using reinforcement learning to shape a soccer

More information

Common Subspace Transfer for Reinforcement Learning Tasks

Common Subspace Transfer for Reinforcement Learning Tasks Common Subspace Transfer for Reinforcement Learning Tasks ABSTRACT Haitham Bou Ammar Institute of Applied Research Ravensburg-Weingarten University of Applied Sciences, Germany bouammha@hs-weingarten.de

More information

Quadruped Robots and Legged Locomotion

Quadruped Robots and Legged Locomotion Quadruped Robots and Legged Locomotion J. Zico Kolter Computer Science Department Stanford University Joint work with Pieter Abbeel, Andrew Ng Why legged robots? 1 Why Legged Robots? There is a need for

More information

Akarsh Pokkunuru EECS Department Contractive Auto-Encoders: Explicit Invariance During Feature Extraction

Akarsh Pokkunuru EECS Department Contractive Auto-Encoders: Explicit Invariance During Feature Extraction Akarsh Pokkunuru EECS Department 03-16-2017 Contractive Auto-Encoders: Explicit Invariance During Feature Extraction 1 AGENDA Introduction to Auto-encoders Types of Auto-encoders Analysis of different

More information

Optimization. Industrial AI Lab.

Optimization. Industrial AI Lab. Optimization Industrial AI Lab. Optimization An important tool in 1) Engineering problem solving and 2) Decision science People optimize Nature optimizes 2 Optimization People optimize (source: http://nautil.us/blog/to-save-drowning-people-ask-yourself-what-would-light-do)

More information

Reinforcement Control via Heuristic Dynamic Programming. K. Wendy Tang and Govardhan Srikant. and

Reinforcement Control via Heuristic Dynamic Programming. K. Wendy Tang and Govardhan Srikant. and Reinforcement Control via Heuristic Dynamic Programming K. Wendy Tang and Govardhan Srikant wtang@ee.sunysb.edu and gsrikant@ee.sunysb.edu Department of Electrical Engineering SUNY at Stony Brook, Stony

More information

Clustering with Reinforcement Learning

Clustering with Reinforcement Learning Clustering with Reinforcement Learning Wesam Barbakh and Colin Fyfe, The University of Paisley, Scotland. email:wesam.barbakh,colin.fyfe@paisley.ac.uk Abstract We show how a previously derived method of

More information

Robotic Search & Rescue via Online Multi-task Reinforcement Learning

Robotic Search & Rescue via Online Multi-task Reinforcement Learning Lisa Lee Department of Mathematics, Princeton University, Princeton, NJ 08544, USA Advisor: Dr. Eric Eaton Mentors: Dr. Haitham Bou Ammar, Christopher Clingerman GRASP Laboratory, University of Pennsylvania,

More information

Combining Deep Reinforcement Learning and Safety Based Control for Autonomous Driving

Combining Deep Reinforcement Learning and Safety Based Control for Autonomous Driving Combining Deep Reinforcement Learning and Safety Based Control for Autonomous Driving Xi Xiong Jianqiang Wang Fang Zhang Keqiang Li State Key Laboratory of Automotive Safety and Energy, Tsinghua University

More information

Efficient Tuning of SVM Hyperparameters Using Radius/Margin Bound and Iterative Algorithms

Efficient Tuning of SVM Hyperparameters Using Radius/Margin Bound and Iterative Algorithms IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 13, NO. 5, SEPTEMBER 2002 1225 Efficient Tuning of SVM Hyperparameters Using Radius/Margin Bound and Iterative Algorithms S. Sathiya Keerthi Abstract This paper

More information

Sample-Efficient Reinforcement Learning for Walking Robots

Sample-Efficient Reinforcement Learning for Walking Robots Sample-Efficient Reinforcement Learning for Walking Robots B. Vennemann Delft Robotics Institute Sample-Efficient Reinforcement Learning for Walking Robots For the degree of Master of Science in Mechanical

More information

4.12 Generalization. In back-propagation learning, as many training examples as possible are typically used.

4.12 Generalization. In back-propagation learning, as many training examples as possible are typically used. 1 4.12 Generalization In back-propagation learning, as many training examples as possible are typically used. It is hoped that the network so designed generalizes well. A network generalizes well when

More information

Policy gradient methods with model predictive control applied to ball bouncing

Policy gradient methods with model predictive control applied to ball bouncing Policy gradient methods with model predictive control applied to ball bouncing Paul Kulchenko Department of Computer Science and Engineering University of Washington, Seattle, WA Email: paul@kulchenko.com

More information

Reinforcement Learning of Walking Behavior for a Four-Legged Robot

Reinforcement Learning of Walking Behavior for a Four-Legged Robot TuM01-3 Reinforcement Learning of Walking Behavior for a Four-Legged Robot Hajime Kimura GEN~FE.DIS.TITECH.AC.JP Toru Yamashita YAMA~FE.DIS.TITECH.AC.JP Shigenobu Kobayashi KOBAYASI~DIS.TITECH.AC.JP Tokyo

More information

Robot learning for ball bouncing

Robot learning for ball bouncing Robot learning for ball bouncing Denny Dittmar Denny.Dittmar@stud.tu-darmstadt.de Bernhard Koch Bernhard.Koch@stud.tu-darmstadt.de Abstract For robots automatically learning to solve a given task is still

More information

Today. Gradient descent for minimization of functions of real variables. Multi-dimensional scaling. Self-organizing maps

Today. Gradient descent for minimization of functions of real variables. Multi-dimensional scaling. Self-organizing maps Today Gradient descent for minimization of functions of real variables. Multi-dimensional scaling Self-organizing maps Gradient Descent Derivatives Consider function f(x) : R R. The derivative w.r.t. x

More information

0.1. alpha(n): learning rate

0.1. alpha(n): learning rate Least-Squares Temporal Dierence Learning Justin A. Boyan NASA Ames Research Center Moett Field, CA 94035 jboyan@arc.nasa.gov Abstract TD() is a popular family of algorithms for approximate policy evaluation

More information

Locally Weighted Learning

Locally Weighted Learning Locally Weighted Learning Peter Englert Department of Computer Science TU Darmstadt englert.peter@gmx.de Abstract Locally Weighted Learning is a class of function approximation techniques, where a prediction

More information

Character Recognition Using Convolutional Neural Networks

Character Recognition Using Convolutional Neural Networks Character Recognition Using Convolutional Neural Networks David Bouchain Seminar Statistical Learning Theory University of Ulm, Germany Institute for Neural Information Processing Winter 2006/2007 Abstract

More information

ICRA 2012 Tutorial on Reinforcement Learning I. Introduction

ICRA 2012 Tutorial on Reinforcement Learning I. Introduction ICRA 2012 Tutorial on Reinforcement Learning I. Introduction Pieter Abbeel UC Berkeley Jan Peters TU Darmstadt Motivational Example: Helicopter Control Unstable Nonlinear Complicated dynamics Air flow

More information

Visual object classification by sparse convolutional neural networks

Visual object classification by sparse convolutional neural networks Visual object classification by sparse convolutional neural networks Alexander Gepperth 1 1- Ruhr-Universität Bochum - Institute for Neural Dynamics Universitätsstraße 150, 44801 Bochum - Germany Abstract.

More information

Research on time optimal trajectory planning of 7-DOF manipulator based on genetic algorithm

Research on time optimal trajectory planning of 7-DOF manipulator based on genetic algorithm Acta Technica 61, No. 4A/2016, 189 200 c 2017 Institute of Thermomechanics CAS, v.v.i. Research on time optimal trajectory planning of 7-DOF manipulator based on genetic algorithm Jianrong Bu 1, Junyan

More information

Planning and Control: Markov Decision Processes

Planning and Control: Markov Decision Processes CSE-571 AI-based Mobile Robotics Planning and Control: Markov Decision Processes Planning Static vs. Dynamic Predictable vs. Unpredictable Fully vs. Partially Observable Perfect vs. Noisy Environment What

More information

Module 1 Lecture Notes 2. Optimization Problem and Model Formulation

Module 1 Lecture Notes 2. Optimization Problem and Model Formulation Optimization Methods: Introduction and Basic concepts 1 Module 1 Lecture Notes 2 Optimization Problem and Model Formulation Introduction In the previous lecture we studied the evolution of optimization

More information

Can Active Memory Replace Attention?

Can Active Memory Replace Attention? Google Brain NIPS 2016 Presenter: Chao Jiang NIPS 2016 Presenter: Chao Jiang 1 / Outline 1 Introduction 2 Active Memory 3 Step by Step to Neural GPU 4 Another two steps: 1. the Markovian Neural GPU 5 Another

More information

Perceptron: This is convolution!

Perceptron: This is convolution! Perceptron: This is convolution! v v v Shared weights v Filter = local perceptron. Also called kernel. By pooling responses at different locations, we gain robustness to the exact spatial location of image

More information

IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 10, NO. 6, NOVEMBER Inverting Feedforward Neural Networks Using Linear and Nonlinear Programming

IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 10, NO. 6, NOVEMBER Inverting Feedforward Neural Networks Using Linear and Nonlinear Programming IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 10, NO. 6, NOVEMBER 1999 1271 Inverting Feedforward Neural Networks Using Linear and Nonlinear Programming Bao-Liang Lu, Member, IEEE, Hajime Kita, and Yoshikazu

More information

Logistic Regression

Logistic Regression Logistic Regression ddebarr@uw.edu 2016-05-26 Agenda Model Specification Model Fitting Bayesian Logistic Regression Online Learning and Stochastic Optimization Generative versus Discriminative Classifiers

More information

Learning of a ball-in-a-cup playing robot

Learning of a ball-in-a-cup playing robot Learning of a ball-in-a-cup playing robot Bojan Nemec, Matej Zorko, Leon Žlajpah Robotics Laboratory, Jožef Stefan Institute Jamova 39, 1001 Ljubljana, Slovenia E-mail: bojannemec@ijssi Abstract In the

More information

Hierarchical Reinforcement Learning for Robot Navigation

Hierarchical Reinforcement Learning for Robot Navigation Hierarchical Reinforcement Learning for Robot Navigation B. Bischoff 1, D. Nguyen-Tuong 1,I-H.Lee 1, F. Streichert 1 and A. Knoll 2 1- Robert Bosch GmbH - Corporate Research Robert-Bosch-Str. 2, 71701

More information

Inverse KKT Motion Optimization: A Newton Method to Efficiently Extract Task Spaces and Cost Parameters from Demonstrations

Inverse KKT Motion Optimization: A Newton Method to Efficiently Extract Task Spaces and Cost Parameters from Demonstrations Inverse KKT Motion Optimization: A Newton Method to Efficiently Extract Task Spaces and Cost Parameters from Demonstrations Peter Englert Machine Learning and Robotics Lab Universität Stuttgart Germany

More information

A fast point-based algorithm for POMDPs

A fast point-based algorithm for POMDPs A fast point-based algorithm for POMDPs Nikos lassis Matthijs T. J. Spaan Informatics Institute, Faculty of Science, University of Amsterdam Kruislaan 43, 198 SJ Amsterdam, The Netherlands {vlassis,mtjspaan}@science.uva.nl

More information

Ensemble methods in machine learning. Example. Neural networks. Neural networks

Ensemble methods in machine learning. Example. Neural networks. Neural networks Ensemble methods in machine learning Bootstrap aggregating (bagging) train an ensemble of models based on randomly resampled versions of the training set, then take a majority vote Example What if you

More information

ARTICLE IN PRESS Neural Networks ( )

ARTICLE IN PRESS Neural Networks ( ) Neural Networks ( ) Contents lists available at ScienceDirect Neural Networks journal homepage: www.elsevier.com/locate/neunet 2008 Special Issue Reinforcement learning of motor skills with policy gradients

More information

Model learning for robot control: a survey

Model learning for robot control: a survey Model learning for robot control: a survey Duy Nguyen-Tuong, Jan Peters 2011 Presented by Evan Beachly 1 Motivation Robots that can learn how their motors move their body Complexity Unanticipated Environments

More information

5 Machine Learning Abstractions and Numerical Optimization

5 Machine Learning Abstractions and Numerical Optimization Machine Learning Abstractions and Numerical Optimization 25 5 Machine Learning Abstractions and Numerical Optimization ML ABSTRACTIONS [some meta comments on machine learning] [When you write a large computer

More information

Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems

Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems Blaise Thomson and Steve Young University of Cambridge Abstract This paper describes a statistically motivated framework

More information

Simulated Transfer Learning Through Deep Reinforcement Learning

Simulated Transfer Learning Through Deep Reinforcement Learning Through Deep Reinforcement Learning William Doan Griffin Jarmin WILLRD9@VT.EDU GAJARMIN@VT.EDU Abstract This paper encapsulates the use reinforcement learning on raw images provided by a simulation to

More information

Notes on Multilayer, Feedforward Neural Networks

Notes on Multilayer, Feedforward Neural Networks Notes on Multilayer, Feedforward Neural Networks CS425/528: Machine Learning Fall 2012 Prepared by: Lynne E. Parker [Material in these notes was gleaned from various sources, including E. Alpaydin s book

More information

Learning Parameterized Skills

Learning Parameterized Skills Bruno Castro da Silva bsilva@cs.umass.edu Autonomous Learning Laboratory, Computer Science Dept., University of Massachusetts Amherst, 13 USA. George Konidaris gdk@csail.mit.edu MIT Computer Science and

More information

Reinforcement learning for imitating constrained reaching movements

Reinforcement learning for imitating constrained reaching movements Advanced Robotics, Vol. 21, No. 13, pp. 1521 1544 (2007) VSP and Robotics Society of Japan 2007. Also available online - www.brill.nl/ar Full paper Reinforcement learning for imitating constrained reaching

More information

Data Mining. Neural Networks

Data Mining. Neural Networks Data Mining Neural Networks Goals for this Unit Basic understanding of Neural Networks and how they work Ability to use Neural Networks to solve real problems Understand when neural networks may be most

More information

Gradient Reinforcement Learning of POMDP Policy Graphs

Gradient Reinforcement Learning of POMDP Policy Graphs 1 Gradient Reinforcement Learning of POMDP Policy Graphs Douglas Aberdeen Research School of Information Science and Engineering Australian National University Jonathan Baxter WhizBang! Labs July 23, 2001

More information

Machine Learning Basics: Stochastic Gradient Descent. Sargur N. Srihari

Machine Learning Basics: Stochastic Gradient Descent. Sargur N. Srihari Machine Learning Basics: Stochastic Gradient Descent Sargur N. srihari@cedar.buffalo.edu 1 Topics 1. Learning Algorithms 2. Capacity, Overfitting and Underfitting 3. Hyperparameters and Validation Sets

More information

A Framework for adaptive focused web crawling and information retrieval using genetic algorithms

A Framework for adaptive focused web crawling and information retrieval using genetic algorithms A Framework for adaptive focused web crawling and information retrieval using genetic algorithms Kevin Sebastian Dept of Computer Science, BITS Pilani kevseb1993@gmail.com 1 Abstract The web is undeniably

More information

Image Registration Lecture 4: First Examples

Image Registration Lecture 4: First Examples Image Registration Lecture 4: First Examples Prof. Charlene Tsai Outline Example Intensity-based registration SSD error function Image mapping Function minimization: Gradient descent Derivative calculation

More information

CHAPTER VI BACK PROPAGATION ALGORITHM

CHAPTER VI BACK PROPAGATION ALGORITHM 6.1 Introduction CHAPTER VI BACK PROPAGATION ALGORITHM In the previous chapter, we analysed that multiple layer perceptrons are effectively applied to handle tricky problems if trained with a vastly accepted

More information

Univariate and Multivariate Decision Trees

Univariate and Multivariate Decision Trees Univariate and Multivariate Decision Trees Olcay Taner Yıldız and Ethem Alpaydın Department of Computer Engineering Boğaziçi University İstanbul 80815 Turkey Abstract. Univariate decision trees at each

More information

Supervised Learning with Neural Networks. We now look at how an agent might learn to solve a general problem by seeing examples.

Supervised Learning with Neural Networks. We now look at how an agent might learn to solve a general problem by seeing examples. Supervised Learning with Neural Networks We now look at how an agent might learn to solve a general problem by seeing examples. Aims: to present an outline of supervised learning as part of AI; to introduce

More information

Dr. Qadri Hamarsheh Supervised Learning in Neural Networks (Part 1) learning algorithm Δwkj wkj Theoretically practically

Dr. Qadri Hamarsheh Supervised Learning in Neural Networks (Part 1) learning algorithm Δwkj wkj Theoretically practically Supervised Learning in Neural Networks (Part 1) A prescribed set of well-defined rules for the solution of a learning problem is called a learning algorithm. Variety of learning algorithms are existing,

More information

IN recent years, neural networks have attracted considerable attention

IN recent years, neural networks have attracted considerable attention Multilayer Perceptron: Architecture Optimization and Training Hassan Ramchoun, Mohammed Amine Janati Idrissi, Youssef Ghanou, Mohamed Ettaouil Modeling and Scientific Computing Laboratory, Faculty of Science

More information

REINFORCEMENT LEARNING WITH HIGH-DIMENSIONAL, CONTINUOUS ACTIONS

REINFORCEMENT LEARNING WITH HIGH-DIMENSIONAL, CONTINUOUS ACTIONS REINFORCEMENT LEARNING WITH HIGH-DIMENSIONAL, CONTINUOUS ACTIONS Leemon C. Baird III and A. Harry Klopf Technical Report WL-TR-93-1147 Wright Laboratory Wright-Patterson Air Force Base, OH 45433-7301 Address:

More information

Gossip Learning. Márk Jelasity

Gossip Learning. Márk Jelasity Gossip Learning Márk Jelasity 2 3 Motivation Explosive growth of smart phone platforms, and Availability of sensor and other contextual data Makes collaborative data mining possible Health care: following

More information

Continuous Valued Q-learning for Vision-Guided Behavior Acquisition

Continuous Valued Q-learning for Vision-Guided Behavior Acquisition Continuous Valued Q-learning for Vision-Guided Behavior Acquisition Yasutake Takahashi, Masanori Takeda, and Minoru Asada Dept. of Adaptive Machine Systems Graduate School of Engineering Osaka University

More information

Performance Comparison of Sarsa(λ) and Watkin s Q(λ) Algorithms

Performance Comparison of Sarsa(λ) and Watkin s Q(λ) Algorithms Performance Comparison of Sarsa(λ) and Watkin s Q(λ) Algorithms Karan M. Gupta Department of Computer Science Texas Tech University Lubbock, TX 7949-314 gupta@cs.ttu.edu Abstract This paper presents a

More information

CS 179 Lecture 16. Logistic Regression & Parallel SGD

CS 179 Lecture 16. Logistic Regression & Parallel SGD CS 179 Lecture 16 Logistic Regression & Parallel SGD 1 Outline logistic regression (stochastic) gradient descent parallelizing SGD for neural nets (with emphasis on Google s distributed neural net implementation)

More information

Predict the box office of US movies

Predict the box office of US movies Predict the box office of US movies Group members: Hanqing Ma, Jin Sun, Zeyu Zhang 1. Introduction Our task is to predict the box office of the upcoming movies using the properties of the movies, such

More information