Skip to content

Translation in progress Some entries still include text in the original language.

Oct 12 · Monday

Reward hacking · Papers

Found 1 paper
Filter1

System type

Models and APIs11AI Agent15coding agent7science agent3Embodied systems and robotics1
Select any value within a category; different categories must all match. Counts include the other filters you have selected.
System type：Embodied systems and roboticsTopic: Reward hackingClear all
  1. Analysis and theoryarXiv 2610.03196

    研究：人形机器人学习流程四层代理均存在偏离，不止奖励作弊

    论证人形机器人学习四层代理可在目标缺失时报成功

    arXiv
    2610.03196
    Published
    Layer
    Training and data layer
    Setting
    Embodied systems and robotics
    Evaluated models
    Gaussian MLP（三隐层各 128 ELU，独立 critic，PPO）
    Abstract四层都可能在目标缺失时报告成功。

    作者讨论腿式机器人 RL 流水线中，奖励、课程门控、评测统计量和参考运动如何在物理目标缺失时仍报告成功。传统观点只把奖励当作会被优化、因而会偏离的代理。

    Paper analysisRead analysis
You have reached the end