Skip to content
Back to skills

Qensemble For Offline Rl Dont Scale The Ensemble Scale The Batch Size

ASecurity

**arXiv ID:** 2211.11092 **Authors:** Alexander Nikulin, Vladislav Kurenkov, Denis Tarasov, Dmitry Akimov, Sergey Kolesnikov **Published:** 2022-11-20T21:48:25Z **Abstract:** Training large neural networks is known to be time-consuming, with the learning duration taking days or even weeks. To address this problem, large-batch optimization was introduced. This approach demonstrated that scaling mini-batch sizes with appropriate learning rate adjustments can speed up the training process by ord...

  • 3 stars
  • 0 votes
  • 0 copies
  • 4 views
  • Added September 11, 2026
toolsgoperformance

Security analysis

A100/100

Scanned September 11, 2026

npx -y skills add hiyenwong/ai_collection --skill qensemble-for-offline-rl-dont-scale-the-ensemble-scale-the-batch-size --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Qensemble For Offline Rl Dont Scale The Ensemble Scale The Batch Size?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Qensemble For Offline Rl Dont Scale The Ensemble Scale The Batch Size
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/hiyenwong-qensemble-for-offline-rl-dont-scale-the-ensemble-s/badge)](https://www.skillsdirectory.com/skills/hiyenwong-qensemble-for-offline-rl-dont-scale-the-ensemble-s)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
# Q-Ensemble for Offline RL: Don't Scale the Ensemble, Scale the Batch Size

**arXiv ID:** 2211.11092
**Authors:** Alexander Nikulin, Vladislav Kurenkov, Denis Tarasov, Dmitry Akimov, Sergey Kolesnikov
**Published:** 2022-11-20T21:48:25Z
**Abstract:**
Training large neural networks is known to be time-consuming, with the learning duration taking days or even weeks. To address this problem, large-batch optimization was introduced. This approach demonstrated that scaling mini-batch sizes with appropriate learning rate adjustments can speed up the training process by orders of magnitude. While long training time was not typically a major issue for model-free deep offline RL algorithms, recently introduced Q-ensemble methods achieving state-of-the-art performance made this issue more relevant, notably extending the training duration. In this work, we demonstrate how this class of methods can benefit from large-batch optimization, which is commonly overlooked by the deep offline RL community. We show that scaling the mini-batch size and naively adjusting the learning rate allows for (1) a reduced size of the Q-ensemble, (2) stronger penalization of out-of-distribution actions, and (3) improved convergence time, effectively shortening training duration by 3-4x times on average.

## Skill Description

This skill is generated from the arXiv paper: Q-Ensemble for Offline RL: Don't Scale the Ensemble, Scale the Batch Size (2211.11092).

## How to Use

[To be filled in by the user or by future automation]

## References

- [arXiv:2211.11092](http://arxiv.org/abs/2211.11092v2)

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in tools →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.