Meta Reinforcement Learning Chelsea Finn Why are humans so good at - PowerPoint PPT Presentation

Jan 06, 2024 •615 likes •1.04k views

Meta Reinforcement Learning Chelsea Finn Why are humans so good at RL? People have prior experience. People have an existing representation of the world. Can we learn a representation under which RL is fast? Key idea : Explicitly optimize for

Meta Reinforcement Learning Chelsea Finn
Why are humans so good at RL? People have prior experience. People have an existing representation of the world. Can we learn a representation under which RL is fast? Key idea : Explicitly optimize for such a representation “Learn how to reinforcement learn”
Outline Meta-RL Problem Formulation & Examples Method Classes: Recurrent Models, Gradient-Based Models Challenges & Latest Developments
The Meta-Learning Problem Supervised Learning: Inputs: Outputs: Data: Meta Supervised Learning: Inputs: Outputs: Data: { Why is this view useful? Reduces the problem to the design & optimization of f . Finn & Levine. Meta-learning and Universality: Deep Representation… ICLR 2018
Example: Few-Shot Classification Given 1 example of 5 classes: Classify new examples test set training data meta-training training classes … … diagram adapted from Ravi & Larochelle ‘17
Meta-RL Example: Maze Navigation Given a small amount of experience Learn to solve the task By learning how to learn many other tasks: … diagram adapted from Duan et al. ‘17
The Meta Reinforcement Learning Problem Reinforcement Learning: Inputs: Outputs: Data: Meta Reinforcement Learning: Inputs: Outputs: Data: { dataset of datasets k rollouts from collected for each task *and* collecting appropriate data Design & optimization of f (learning to explore) Finn. Learning to Learn with Gradients. PhD Thesis 2018
Meta-RL Example: Maze Navigation Given a small amount of experience Learn to solve the task [meta] test time [meta] train time By learning how to learn many other tasks: meta-training … tasks diagram adapted from Duan et al. ‘17
The Meta Reinforcement Learning Problem Meta Reinforcement Learning: Inputs: Outputs: { Episodic Variant k rollouts from Inputs: Outputs: { Online Variant 1…k timesteps from
Outline Meta-RL Problem Formulation & Examples Method Classes: Recurrent Models, Gradient-Based Models Challenges & Latest Developments

Download Presentation

Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Meta Reinforcement Learning as Task Inference Jan Humplik, Alexandre Galashov, Leonard

Meta Reinforcement Learning as Task Inference Jan Humplik, Alexandre Galashov, Leonard Hasenclever, Pedro A.Ortega, Yee Whye Teh, Nicholas Heess Topic: Bayesian RL Presenter: Ram Ananth Why meta Reinforcement Learning? First Wave of Deep

912 views • 63 slides

Meta Reinforcement Learning Kate Rakelly 11/13/19 Questions we seek to answer Motivation : What

Meta Reinforcement Learning Kate Rakelly 11/13/19 Questions we seek to answer Motivation : What problem is meta-RL trying to solve? Context : What is the connection to other problems in RL? Solutions : What are solution methods for meta-RL and

787 views • 52 slides

Prefrontal cortex as a meta-reinforcement learning system Wang et al. CS330 Student

Prefrontal cortex as a meta-reinforcement learning system Wang et al. CS330 Student Presentation Motivation Computational Neuro: AI <> Neurobio Feedback Loop Convolutions and the eye, SNNs and Learning Rules, etc. Meta

283 views • 16 slides

Reinforcement Learning by the People and for the People: With a Focus on Lifelong / Meta /

Reinforcement Learning by the People and for the People: With a Focus on Lifelong / Meta / Transfer Learning Emma Brunskill Stanford CS234 Winter 2018 Quiz Information Monday, in class See piazza for room information (Released by

942 views • 81 slides

Learning a Prior over Intent via Meta-Inverse Reinforcement Learning Kelvin Xu, Ellis Ratner, Anca

Learning a Prior over Intent via Meta-Inverse Reinforcement Learning Kelvin Xu, Ellis Ratner, Anca Dragan, Sergey Levine, Chelsea Finn University of California, Berkeley Motivation : a well specified reward function remains an important

924 views • 42 slides

Distributed Meta Optimization of Reinforcement Learning Agents Greg Heinrich, Iuri Frosio - GTC

Distributed Meta Optimization of Reinforcement Learning Agents Greg Heinrich, Iuri Frosio - GTC San Jose, March 2019 Contents Introduction to Reinforcement Learning Introduction to Metaoptimization (on distributed systems) / Maglev

596 views • 45 slides

Efficient Off-Policy Meta- Reinforcement Learning via Probabilistic Context Variables Rakelly,

Efficient Off-Policy Meta- Reinforcement Learning via Probabilistic Context Variables Rakelly, K., Zhou, A., Quillen, D., Finn, C., & Levine, S ICML, 2019 Presented by: Egill Ian Gudmundsson Some Terminology On-policy learning: Only one

427 views • 16 slides

PEARL Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables Kate

PEARL Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables Kate Rakelly*, Aurick Zhou*, Deirdre Quillen, Chelsea Finn, Sergey Levine Hula Beach, Never grow up, The Sled - by artist Matt Spangler,

603 views • 12 slides

Unsupervised Meta-Learning for Reinforcement Learning LAMDA, Nanjing University . .

. . . . . . . . . . . . . . . . . Unsupervised Meta-Learning for Reinforcement Learning LAMDA, Nanjing University . . . . . . . . . . . . . . . . . . . . . . . November 9, 2020 . . . . . .

664 views • 47 slides

Causal Reasoning from Meta-reinforcement Learning Dasgupta et al. (2018) CS330 Student

Causal Reasoning from Meta-reinforcement Learning Dasgupta et al. (2018) CS330 Student Presentation Background: Why Causal Reasoning? There is only so much of the world we can understand via observation. Cancer (correlates to) Smoking

297 views • 19 slides

RL Overview of topics About Reinforcement Learning The Reinforcement Learning Problem

Introduction to Reinforcement Learning RL Overview of topics About Reinforcement Learning The Reinforcement Learning Problem Inside an RL agent Temporal difference learning Many faces of Reinforcement Learning What is

551 views • 35 slides

Prefrontal cortex as a Meta-reinforcement learning system Matthew Botvinick DeepMind, London UK

Prefrontal cortex as a Meta-reinforcement learning system Matthew Botvinick DeepMind, London UK Gatsby Computational Neuroscience Unit, UCL Mnih et al, Nature (2015) Mnih et al, Nature (2015) Yamins & DiCarlo, 2016 Schultz et al, Science

535 views • 49 slides

Meta Learning Shengchao Liu Background Meta Learning (AKA Learning to Learn) A

Meta Learning Shengchao Liu Background Meta Learning (AKA Learning to Learn) A fast-learning algorithm: quickly adapted from the source tasks to the target tasks Key terminologies Support Set & Query Set C-Way K-Shot

2k views • 41 slides

Reinforcement Learning Reinforcement Learning Reinforcement Learning in a nutshell g Imagine

Reinforcement Learning Reinforcement Learning Reinforcement Learning in a nutshell g Imagine playing a new game whose rules you dont know; after a hundred or so moves your don t know; after a hundred or so moves, your opponent announces, You

511 views • 30 slides

Reinforcement Learning You can think of supervised learning as the teacher providing answers

Reinforcement Learning You can think of supervised learning as the teacher providing answers (the class labels) Reinforcement learning: In reinforcement learning, the agent learns Markov Decision Processes based on a punishment/reward

89 views • 6 slides

Examples of Reinforcement Learning Robocup Soccer Teams Stone & Veloso, Reidmiller et al.

Reinforcement Learning Slides by Rich Sutton Mods by Dan Lizotte Refer to Reinforcement Learning: An Introduction by Sutton and Barto Alpaydin Chapter 16 Up until now we have been Supervised Learning Classifying, mostly

891 views • 31 slides

Slides for Lecture 27 ENEL 353: Digital Circuits Fall 2013 Term Steve Norman, PhD, PEng

Slides for Lecture 27 ENEL 353: Digital Circuits Fall 2013 Term Steve Norman, PhD, PEng Electrical & Computer Engineering Schulich School of Engineering University of Calgary 13 November, 2013 slide 2/19 ENEL 353 F13 Section 02

333 views • 19 slides

D Answer Answer D .003 m/s D .003 m/s [This object is a pull tab] Slide 4 / 38 Slide 4

Slide 1 / 38 Slide 2 / 38 New Jersey Center for Teaching and Learning Progressive Science Initiative Forces & Motion This material is made freely available at www.njctl.org and is intended for the non-commercial use of students and

494 views • 13 slides

Matsya Systems Next generation Underwater Ship Hull Cleaning Customer Pain Points Biofouling

Matsya Systems Next generation Underwater Ship Hull Cleaning Customer Pain Points Biofouling occurs on ships, despite advances like anti-biofouling paint Causes increase in fuel consumption by 5-10% (mild growth) 10-25% (within 4-6 months)

1.19k views • 63 slides

Evolutionary Search Knowledge Reservoir Set of possible solutions Gleaning a reservoir of

global maximum local maxima local maxima Evolutionary Search Knowledge Reservoir Set of possible solutions Gleaning a reservoir of knowledge om interactions ith the environment. Selection Fitness dependent number of o

483 views • 8 slides

Snail Express Conquering the Chaos of Mail User Feedback Retrieving Mail = Easy Opening

Red B Snail Express Conquering the Chaos of Mail User Feedback Retrieving Mail = Easy Opening Mail = 50% Unopened Bills and Clutter = Stress How It Works Scan Open Extract Unfold Scan Extract Vacuum + Shake

473 views • 10 slides

Snail deterrent properties of a soot based flexible superhydrophobic surface Nicasio Geraldi,

Snail deterrent properties of a soot based flexible superhydrophobic surface Nicasio Geraldi, Robert Morris, Glen McHale and Michael Newton Introduction Snails enjoying eating the leaves of many garden plants including food crops.

425 views • 13 slides

2020 MS. KELLYS SIXTH GRADE GLOBAL THINKERS STUDENT OF THE WEEK: STUDENT OF THE WEEK FOR

INSTRUCTION WEEK OF MAY 4 TH 2020 MS. KELLYS SIXTH GRADE GLOBAL THINKERS STUDENT OF THE WEEK: STUDENT OF THE WEEK FOR APRIL 27 TH 2020: SAMANTA! FOR ALWAYS ATTENDING ZOOM SESSIONS AND WORKING HARD TO SUBMIT WORK EACH DAY LAST WEEK!

763 views • 30 slides

D Answer 600 m/s C 0.003 m/s D [This object is a pull tab] Slide 4 / 40 2 A blimp travels

Slide 1 / 40 Slide 2 / 40 8th Grade Forces and Motion Study Guide www.njctl.org Slide 3 / 40 1 A snail travels 10 m in 3000 seconds. What is the snail's average speed? 60000 m/s A 0.02 m/s B 600 m/s C 0.003 m/s D Slide 3 (Answer) /

474 views • 26 slides