About
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 |
|---|---|---|---|---|---|---|
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
||||||
|
Subtasks
Datasets
Latest papers with code
Revisiting Prioritized Experience Replay: A Value Perspective
Furthermore, we successfully extend our theoretical framework to maximum-entropy RL by deriving the lower and upper bounds of these value metrics for soft Q-learning, which turn out to be the product of $|\text{TD}|$ and "on-policyness" of the experiences.
Shielding Atari Games with Bounded Prescience
• HjalmarWijk/bounded-prescience
• 
We present the first exact method for analysing and ensuring the safety of DRL agents for Atari games.
Benchmarking Perturbation-based Saliency Maps for Explaining Deep Reinforcement Learning Agents
All four approaches work by perturbing parts of the input and measuring how much this affects the agent's output.
Developing an OpenAI Gym-compatible framework and simulation environment for testing Deep Reinforcement Learning agents solving the Ambulance Location Problem
Results: A range of Deep RL agents based on Deep Q networks were tested in this custom environment.
Evolving Reinforcement Learning Algorithms
Learning from scratch on simple classical control and gridworld tasks, our method rediscovers the temporal-difference (TD) algorithm.
Reinforcement Learning with Latent Flow
• WendyShang/flare
• 
Temporal information is essential to learning effective policies with Reinforcement Learning (RL).
CONTINUOUS CONTROL MONTEZUMA'S REVENGE OPTICAL FLOW ESTIMATION VIDEO CLASSIFICATION
Multi-Agent Trust Region Learning
We derive the lower bound of agents' payoff improvements for MATRL methods, and also prove the convergence of our method on the meta-game fixed points.
Augmenting Policy Learning with Routines Discovered from a Demonstration
• sjtuytc/-AAAI21-RoutineAugmentedPolicyLearning-RAPL-
•
Humans can abstract prior knowledge from very little data and use it to boost skill learning.
Evaluating Agents without Rewards
Moreover, input entropy and information gain correlate more strongly with human similarity than task reward does, suggesting the use of intrinsic objectives for designing agents that behave similarly to human players.
High-Throughput Synchronous Deep RL
NeurIPS 2020
• IouJenLiu/HTS-RL
•
In contrast, asynchronous methods achieve high throughput but suffer from stability issues and lower sample efficiency due to `stale policies.'