论文/arXiv 2609.13730·1 个月前

用来自160,000次训练运行的经验启动你的策略学习

为提升离线策略学习可靠性，研究通过16万余次策略训练系统分析算法、调参与基准影响，并在114个数据集上验证。

AI 整理

• 数据与采集
• 具身模型
• 评测与安全

原文链接：JumpStart Your Policy Learning with Lessons from 160,000 Training Runs（arXiv 2609.13730 · 发布日期：2026-09-12）

JumpStart Your Policy Learning with Lessons from 160,000 Training Runs

为提升离线策略学习可靠性，研究通过16万余次策略训练系统分析算法、调参与基准影响，并在114个数据集上验证。

核心要点

• 问题：报告选择、超参数调优和数据集属性会影响结论，但缺少联合的大规模系统考察。
• 方法：训练超过160,000个策略，覆盖114个数据集，分析算法表现、调参敏感性与跨环境迁移。
• 结果：没有算法始终占优，强方法总体接近；调参和基准组成会改变算法排名与结论。
• 发布JumpStart，包含训练策略、分数、超参数、强基线、训练评估代码和可扩展网站。

验证范围与局限：验证范围为离线强化学习和模仿学习的114个数据集。

英文摘要 (Abstract)

Reliable progress in offline policy learning depends on careful reporting, well-tuned baselines, and evaluation across diverse conditions. Prior work has shown that results can be sensitive to reporting choices, hyperparameter tuning, and dataset properties, but these sources of variability have not been systematically investigated together at the scale needed to understand how they shape conclusions. To address this gap, we present a large-scale empirical study of offline reinforcement and imitation learning, training over 160,000 policies across 114 datasets. At this scale, no algorithm dominates: aggregate performance among the strongest methods is often close, but the leaders differ substantially across environments. We find that proper hyperparameter tuning frequently reshuffles perceived algorithm rankings and that benchmark composition can produce conflicting conclusions. We also study hyperparameter sensitivity and transfer across environments, identifying a simple strategy for deriving strong default configurations. We use our findings to develop a dataset-conditioned recommender that provides task-specific algorithm recommendations for practitioners. Finally, we release JumpStart: a resource suite containing every trained policy, per-model scores and hyperparameters, strong baselines across all environments, training and evaluation code, and an extensible website for retrieving, analyzing, and contributing results. Together, these resources aim to make offline policy-learning research more reliable and enable future work beyond the scope of this study.

作者：Nabil Omi, Eric Bae, Chung Yik Edward Yeung, Siddhartha Sen, Ali Farhadi

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2609.13730：JumpStart Your Policy Learning with Lessons from 160,000 Training Runs访问 2026-10-21