解决自适应学习中数据贡献度评估难题，让模型知道每条数据的真实价值。
Data Attribution in Adaptive Learning
- 提出基于条件干预的目标函数，精准刻画自适应学习中每条数据的贡献
- 证明仅靠回放数据无法恢复真实贡献度，除非满足特定结构条件
- 适用于在线学习、强化学习等动态数据场景，适合研究模型可解释性者
机器学习模型正越来越多地生成自己的训练数据——在线博弈、强化学习以及语言模型的后训练流程是典型例子。在这些自适应设置中，单个训练样本不仅更新学习器，还改变未来数据的分布。标准归因方法针对静态数据集设计，忽略了这种反馈机制。本文通过条件干预目标，形式化了有限时域自适应学习中的出现级归因；证明了仅依赖回放侧信息一般无法恢复该归因；并识别出一类结构，在该类下可从日志数据中唯一确定归因目标。
原文摘要 · Abstract (English)
Machine learning models increasingly generate their own training data -- online bandits, reinforcement learning, and post-training pipelines for language models are leading examples. In these adaptive settings, a single training observation both updates the learner and shifts the distribution of future data the learner will collect. Standard attribution methods, designed for static datasets, ignore this feedback. We formalize occurrence-level attribution for finite-horizon adaptive learning via a conditional interventional target, prove that replay-side information cannot recover it in general, and identify a structural class in which the target is identified from logged data.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。