当验证阻碍学习：面向持续具身智能体的更新准入审计
When Validation Stops Learning: Auditing Update Admission for Continual Embodied Agents
浏览论文内容
中文总结 AI 辅助
本文提出一种更新准入审计协议，通过错误控制与学习机会双重评估，利用配对二项式检验等方法来平衡持续具身智能体的安全性与学习能力。
中文摘要 AI 辅助
独立评估能够拒绝有害的策略更新，但也可能阻碍有益的持续学习。我们认为，更新准入必须通过错误控制以及在既定交互预算内保留的学习机会来共同评估。我们识别出一个具体缺陷：基于范围的置信门在原本充足的预算内无法证明旧任务行为不变。当结果分歧罕见时，标准的配对二项式构造可减轻这一负担。我们还明确了经认证的历史参考提升以及回合级错失机会指标。在一个构造的单步推动诊断中（32个随机种子），新鲜的配对检查在每阶段2000个回合下接纳了常见更新流的31.6%，而基于范围的门的接纳率为零；然而，在闭环运行中，无条件回放的学习效果更好。一项单独的学得动力学压力测试区分了模型偏差与反馈选择误差。我们的贡献在于提出了一套结合分析性与合成性证据的准入审计协议；物理机器人和VLA验证仍有待开展。
英文摘要
Independent evaluation can reject harmful policy updates yet also prevent useful continual learning. We argue that update admission must be assessed through both error control and retained learning opportunities at a stated interaction budget. We identify a concrete failure: a range-based confidence gate cannot certify unchanged old-task behavior within otherwise substantial budgets. A standard paired-binomial construction reduces this burden when outcome disagreements are rare. We also specify certified historical-reference promotion and a round-level missed-opportunity metric. In a constructed one-step pushing diagnostic with 32 seeds, fresh paired checks admit 31.6% of a common update stream at 2,000 episodes per stage, versus zero for the range-based gate; unconditional replay nevertheless learns better in closed-loop runs. A separate learned-dynamics stress test distinguishes model bias from feedback-selection error. The contribution is an admission-audit protocol with analytical and synthetic evidence; physical-robot and VLA validation remain open.
发表机构
- Rice University（莱斯大学）
- University of California, Berkeley（加州大学伯克利分校）
机构由 AI 辅助整理，请以论文原文为准。
