Pith. sign in

REVIEW 3 cited by

Is Deep Reinforcement Learning Really Superhuman on Atari? Leveling the playing field

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1908.04683 v5 pith:FDHQZNYP submitted 2019-08-13 cs.AI

classification cs.AI
keywords learningenvironmentperformancereinforcementatarideepdifferentintroduce
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Consistent and reproducible evaluation of Deep Reinforcement Learning (DRL) is not straightforward. In the Arcade Learning Environment (ALE), small changes in environment parameters such as stochasticity or the maximum allowed play time can lead to very different performance. In this work, we discuss the difficulties of comparing different agents trained on ALE. In order to take a step further towards reproducible and comparable DRL, we introduce SABER, a Standardized Atari BEnchmark for general Reinforcement learning algorithms. Our methodology extends previous recommendations and contains a complete set of environment parameters as well as train and test procedures. We then use SABER to evaluate the current state of the art, Rainbow. Furthermore, we introduce a human world records baseline, and argue that previous claims of expert or superhuman performance of DRL might not be accurate. Finally, we propose Rainbow-IQN by extending Rainbow with Implicit Quantile Networks (IQN) leading to new state-of-the-art performance. Source code is available for reproducibility.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Playstyle and Artificial Intelligence: An Initial Blueprint Through the Lens of Video Games

    cs.AI 2025-08 conditional novelty 6.0 of 10

    This dissertation formalizes playstyle as the decision-making style of an agent, introduces a discrete-state playstyle distance that distinguishes behaviors in racing games and Atari, and proposes a blueprint for usin...

  2. Augmenting the action space with conventions to improve multi-agent cooperation in Hanabi

    cs.MA 2024-12 conditional novelty 6.0 of 10

    Agents trained with an action space augmented by human Hanabi conventions learn faster and score higher than baseline Rainbow agents, and they cooperate better with unfamiliar partners.

  3. Decorrelated Soft Actor-Critic for Efficient Deep Reinforcement Learning

    cs.LG 2025-01 conditional novelty 5.0 of 10

    Decorrelated Soft Actor-Critic (DSAC) adds layerwise input decorrelation to discrete SAC and reports faster wall-clock training in 5 of 7 Atari games and better reward in 2, though the gains are partly confounded by p...

Pith tools