arXiv research

Agent57 beats human average on all Atari games.

Proposes a new Deep Recurrent Double Q-Learning model for Atari games.

Model-free reinforcement learning (RL) can be used to learn effective policies for complex tasks, such as Atari games, even from image observations. However, this typically requires very large amounts of interaction -- substantially more, in fact, than a human would need to learn the same games. How can people learn so…

Method learns state representations without supervision for Atari games.

A new RL method boosts exploration efficiency in Atari games and driving simulators.

Develops a deep metric learning approach for detecting bugs in video games.

Despite significant advances in the field of deep Reinforcement Learning (RL), today's algorithms still fail to learn human-level policies consistently over a set of diverse tasks such as Atari 2600 games. We identify three key challenges that any algorithm needs to master in order to perform well on all games: process…

Reinforcement learning is concerned with identifying reward-maximizing behaviour policies in environments that are initially unknown. State-of-the-art reinforcement learning approaches, such as deep Q-networks, are model-free and learn to act effectively across a wide range of environments such as Atari games, but requ…

DreamerV2 learns Atari game behaviors from a world model, achieving human-level performance.

Video games improve vehicle routing performance.

Low precision networks in the reinforcement learning (RL) setting are relatively unexplored because of the limitations of binary activations for function approximation. Here, in the discrete action ATARI domain, we demonstrate, for the first time, that low precision policy distillation from a high precision network pro…

A new RL method improves performance on Atari games without complex techniques.

Deep learning improves reinforcement learning in video games.

AG-RL uses action grammars to improve reinforcement learning efficiency.

To solve complex real-world problems with reinforcement learning, we cannot rely on manually specified reward functions. Instead, we can have humans communicate an objective to the agent directly. In this work, we combine two approaches to learning from human feedback: expert demonstrations and trajectory preferences. …

MuZero learns models to master complex games without domain knowledge.

This work compresses reinforcement learning models for Atari games, improving localization.

Paper presents a new method for Bayesian deep learning that scales to Atari games.

The study explores when parametric models enhance reinforcement learning, validating a hypothesis on Atari games.

Paper proposes a new method for better estimating continuous distributions in RL.

Improved deep Q-learning with SVRG reduces variance and stabilizes training.

New Soft Actor-Critic for discrete actions.

Bayesian REX learns Atari games from demonstrations efficiently.

Paper proposes an RL algorithm to ensure policy performance guarantees.

Deep reinforcement learning has become popular over recent years, showing superiority on different visual-input tasks such as playing Atari games and robot navigation. Although objects are important image elements, few work considers enhancing deep reinforcement learning with object characteristics. In this paper, we p…

Generative deep learning creates counterfactual states to explain Atari agent decisions.

In this work, we build on recent advances in distributional reinforcement learning to give a generally applicable, flexible, and state-of-the-art distributional variant of DQN. We achieve this by using quantile regression to approximate the full quantile function for the state-action return distribution. By reparameter…

Adapting momentum from optimization to reinforcement learning.

Study finds reinforcement learning performance plateaus due to environmental interference.

Achieving efficient and scalable exploration in complex domains poses a major challenge in reinforcement learning. While Bayesian and PAC-MDP approaches to the exploration problem offer strong formal guarantees, they are often impractical in higher dimensions due to their reliance on enumerating the state-action space.…

Scaling up model and data size improves imitation learning in single-agent games.

A new reward learning module improves imitation learning in high-dimensional environments.

Due to the capability of deep learning to perform well in high dimensional problems, deep reinforcement learning agents perform well in challenging tasks such as Atari 2600 games. However, clearly explaining why a certain action is taken by the agent can be as important as the decision itself. Deep reinforcement learni…

In this work, we ask the following question: Can visual analogies, learned in an unsupervised way, be used in order to transfer knowledge between pairs of games and even play one game using an agent trained for another game? We attempt to answer this research question by creating visual analogies between a pair of game…

Large-scale public datasets have been shown to benefit research in multiple areas of modern artificial intelligence. For decision-making research that requires human data, high-quality datasets serve as important benchmarks to facilitate the development of new methods by providing a common reproducible standard. Many h…

CURL uses contrastive learning to improve reinforcement learning performance.

DPFRL uses particle filters for decision making with complex visual observations.

Combines experience replay and exploration for better agent performance.

Method learns statistics of return distributions via neural networks and maximum mean discrepancy.

A new method for reinforcement learning scales errors without tuning.

Deep reinforcement learning, applied to vision-based problems like Atari games, maps pixels directly to actions; internally, the deep neural network bears the responsibility of both extracting useful information and making decisions based on it. By separating the image processing from decision-making, one could better …

ConQUR tackles delusional bias in deep Q-learning, improving performance in Atari games.

New Bellman error estimator improves offline model selection performance.

B-REX efficiently learns Atari game policies from pixel inputs using Bayesian methods.

A new method prioritizes and recycles experiences for better reinforcement learning.

New approach to understand recurrent policies as FSMs without minimization.

Paper proposes Terminal Prediction to improve deep RL performance.

A framework disentangles controllable objects from visual signals for improved RL.