Timezone: »
Deep reinforcement learning methods traditionally struggle with tasks where environment rewards are particularly sparse. One successful method of guiding exploration in these domains is to imitate trajectories provided by a human demonstrator. However, these demonstrations are typically collected under artificial conditions, i.e. with access to the agent’s exact environment setup and the demonstrator’s action and reward trajectories. Here we propose a method that overcomes these limitations in two stages. First, we learn to map unaligned videos from multiple sources to a common representation using self-supervised objectives constructed over both time and modality (i.e. vision and sound). Second, we embed a single YouTube video in this representation to learn a reward function that encourages an agent to imitate human gameplay. This method of one-shot imitation allows our agent to convincingly exceed human-level performance on the infamously hard exploration games Montezuma’s Revenge, Pitfall! and Private Eye for the first time, even if the agent is not presented with any environment rewards.
Author Information
Yusuf Aytar (DeepMind)
Tobias Pfaff (DeepMind)
David Budden (DeepMind)
Thomas Paine (DeepMind)
Ziyu Wang (Deepmind)
Nando de Freitas (DeepMind)
Related Events (a corresponding poster, oral, or spotlight)
-
2018 Poster: Playing hard exploration games by watching YouTube »
Wed Dec 5th through Thu the 6th Room Room 517 AB
More from the Same Authors
-
2020 Poster: Critic Regularized Regression »
Ziyu Wang · Alexander Novikov · Konrad Zolna · Josh Merel · Jost Tobias Springenberg · Scott Reed · Bobak Shahriari · Noah Siegel · Caglar Gulcehre · Nicolas Heess · Nando de Freitas -
2020 Poster: Modular Meta-Learning with Shrinkage »
Yutian Chen · Abram Friesen · Feryal Behbahani · Arnaud Doucet · David Budden · Matthew Hoffman · Nando de Freitas -
2020 Spotlight: Modular Meta-Learning with Shrinkage »
Yutian Chen · Abram Friesen · Feryal Behbahani · Arnaud Doucet · David Budden · Matthew Hoffman · Nando de Freitas -
2020 Poster: A Combinatorial Perspective on Transfer Learning »
Jianan Wang · Eren Sezener · David Budden · Marcus Hutter · Joel Veness -
2020 Poster: RL Unplugged: A Collection of Benchmarks for Offline Reinforcement Learning »
Caglar Gulcehre · Ziyu Wang · Alexander Novikov · Thomas Paine · Sergio Gómez · Konrad Zolna · Rishabh Agarwal · Josh Merel · Daniel Mankowitz · Cosmin Paduraru · Gabriel Dulac-Arnold · Jerry Li · Mohammad Norouzi · Matthew Hoffman · Nicolas Heess · Nando de Freitas -
2020 Poster: Online Learning in Contextual Bandits using Gated Linear Networks »
Eren Sezener · Marcus Hutter · David Budden · Jianan Wang · Joel Veness -
2020 Poster: Gaussian Gated Linear Networks »
David Budden · Adam Marblestone · Eren Sezener · Tor Lattimore · Gregory Wayne · Joel Veness -
2019 Workshop: Science meets Engineering of Deep Learning »
Levent Sagun · Caglar Gulcehre · Adriana Romero · Negar Rostamzadeh · Nando de Freitas -
2019 Poster: Learning Compositional Neural Programs with Recursive Tree Search and Planning »
Thomas PIERROT · Guillaume Ligner · Scott Reed · Olivier Sigaud · Nicolas Perrin · Alexandre Laterre · David Kas · Karim Beguir · Nando de Freitas -
2019 Spotlight: Learning Compositional Neural Programs with Recursive Tree Search and Planning »
Thomas PIERROT · Guillaume Ligner · Scott Reed · Olivier Sigaud · Nicolas Perrin · Alexandre Laterre · David Kas · Karim Beguir · Nando de Freitas -
2017 Poster: Robust Imitation of Diverse Behaviors »
Ziyu Wang · Josh Merel · Scott Reed · Nando de Freitas · Gregory Wayne · Nicolas Heess -
2017 Tutorial: Deep Learning: Practice and Trends »
Nando de Freitas · Scott Reed · Oriol Vinyals -
2016 Workshop: Neural Abstract Machines & Program Induction »
Matko Bošnjak · Nando de Freitas · Tejas Kulkarni · Arvind Neelakantan · Scott E Reed · Sebastian Riedel · Tim Rocktäschel -
2016 Poster: Learning to learn by gradient descent by gradient descent »
Marcin Andrychowicz · Misha Denil · Sergio Gómez · Matthew Hoffman · David Pfau · Tom Schaul · Nando de Freitas -
2015 Workshop: Bayesian Optimization: Scalability and Flexibility »
Bobak Shahriari · Ryan Adams · Nando de Freitas · Amar Shah · Roberto Calandra