Atari Games

126 papers with code ยท Playing Games
Subtask of Video Games

The Atari 2600 Games task (and dataset) involves training an agent to achieve high game scores.

( Image credit: Playing Atari with Deep Reinforcement Learning )

Benchmarks

TREND DATASET BEST METHOD PAPER TITLE PAPER CODE COMPARE

Latest papers without code

Contrastive Variational Model-Based Reinforcement Learning for Complex Observations

6 Aug 2020

This paper presents Contrastive Variational Reinforcement Learning (CVRL), an MBRL framework for complex natural observations.

ATARI GAMES CONTINUOUS CONTROL CONTRASTIVE LEARNING DECISION MAKING

Noisy Agents: Self-supervised Exploration by Predicting Auditory Events

27 Jul 2020

Humans integrate multiple sensory modalities (e. g. visual and audio) to build a causal understanding of the physical world.

ATARI GAMES

Slot Contrastive Networks: A Contrastive Approach for Representing Objects

18 Jul 2020

Unsupervised extraction of objects from low-level visual data is an important goal for further progress in machine learning.

ATARI GAMES

Discovering Reinforcement Learning Algorithms

17 Jul 2020

Automating the discovery of update rules from data could lead to more efficient algorithms, or algorithms that are better adapted to specific environments.

ATARI GAMES META-LEARNING

Analysis of Q-learning with Adaptation and Momentum Restart for Gradient Descent

15 Jul 2020

In this paper, we first characterize the convergence rate for Q-AMSGrad, which is the Q-learning algorithm with AMSGrad update (a commonly adopted alternative of Adam for theoretical analysis).

ATARI GAMES Q-LEARNING

DinerDash Gym: A Benchmark for Policy Learning in High-Dimensional Action Space

13 Jul 2020

It has been arduous to assess the progress of a policy learning algorithm in the domain of hierarchical task with high dimensional action space due to the lack of a commonly accepted benchmark.

ATARI GAMES

Learning Abstract Models for Strategic Exploration and Fast Reward Transfer

12 Jul 2020

Model-based reinforcement learning (RL) is appealing because (i) it enables planning and thus more strategic exploration, and (ii) by decoupling dynamics from rewards, it enables fast transfer to new reward functions.

MONTEZUMA'S REVENGE

Attention or memory? Neurointerpretable agents in space and time

9 Jul 2020

In neuroscience, attention has been shown to bidirectionally interact with reinforcement learning (RL) processes.

ATARI GAMES DIMENSIONALITY REDUCTION

Double Prioritized State Recycled Experience Replay

8 Jul 2020

Experience replay enables online reinforcement learning agents to store and reuse the experiences generated in previous interaction with the environment.

ATARI GAMES

Learning Dialog Policies from Weak Demonstrations

ACL 2020

Deep reinforcement learning is a promising approach to training a dialog manager, but current methods struggle with the large state and action spaces of multi-domain dialog systems.

ATARI GAMES Q-LEARNING