Data Attribution in Adaptive Learning

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteur principal: Rege, Amit Kiran
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913008084582400
author Rege, Amit Kiran
author_facet Rege, Amit Kiran
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2604_04892
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Data Attribution in Adaptive Learning
Rege, Amit Kiran
Machine Learning
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.
title Data Attribution in Adaptive Learning
topic Machine Learning
url https://arxiv.org/abs/2604.04892