跳到正文
返回论文列表
stat.ML提交于 已译

风险最优的主动学习与信息性弃权

Risk-optimal active learning with informative abstention

Fariborz Setoudehtazangi · Geoffrey J. McLachlan

中文摘要

研究当预言机(oracle)可能弃权且其响应概率与分类器共享参数时的标签获取问题。此时一次查询可通过返回的标签和响应结果两方面提供信息。在正确设定的参数模型下,推导出经响应模型冗余参数调整后的分类器信息量。将该信息量与多类零一风险的局部曲率结合,得到最小化渐近超额风险主导项的获取准则。刻画固定查询预算下的最优分配,并展示查询分配如何影响响应可识别性。当响应概率不依赖分类器参数时,该预言机准则退化为响应折扣获取。模拟实验考察所得查询分配,并通过消融隔离响应结果项的贡献。Consumer Complaints 回放保留观察到的人类弃权,并在响应时揭示基准标签。在两个最大预算下,响应感知获取的平均测试误差低于响应折扣获取,配对差异经 Holm 校正后显著。最小预算下熵采样表现更好。增益取决于响应机制和查询预算。

关键要点

  1. 01预言机弃权概率与分类器共享参数时,查询同时通过标签和响应结果提供信息
  2. 02结合调整后分类器信息量与多类零一风险局部曲率,导出最小化渐近超额风险的获取准则
  3. 03固定预算下刻画最优查询分配,并分析其对响应可识别性的影响
  4. 04Consumer Complaints 回放中,两个最大预算下响应感知获取测试误差显著低于响应折扣获取
  5. 05最小预算下熵采样更优,增益依赖响应机制与查询预算

解读

尚无解读。

原始英文摘要

arXiv:2611.24986v1 Announce Type: new Abstract: We study label acquisition when an oracle may abstain and its response probability shares parameters with the classifier. A query can then provide information through both a returned label and the response outcome. Under a correctly specified parametric model, we derive the classifier information after adjustment for response-model nuisance parameters. Combining this information with the local curvature of multiclass zero-one risk gives an acquisition criterion for minimizing the leading asymptotic excess risk. We characterize the optimal allocation under a fixed query budget and show how query allocation affects response identifiability. The oracle criterion reduces to response-discounted acquisition when response probability does not depend on the classifier parameters. Simulations examine the resulting query allocations and isolate the contribution of the response-outcome term through an ablation. A Consumer Complaints replay retains observed human abstentions and reveals benchmark labels upon response. Response-aware acquisition has lower mean test error than response-discounted acquisition at the two largest budgets, with paired differences significant after Holm correction. Entropy sampling performs better at the smallest budget. The gains depend on the response mechanism and query budget.

同方向论文 · stat.ML

查看全部 →