paper-with-me

홈 › Papers

Better Training Data Attribution via Better Inverse Hessian-Vector Products

2025-07-19 · Andrew Wang, Elisa Nguyen, Runshi Yang, Juhan Bae, Sheila A. McIlraith, Roger Grosse arxiv

Training data attribution (TDA) provides insights into which training data is responsible for a learned model behavior. Gradient-based TDA methods such as influence functions and unrolled differentiation both involve a computation that resembles an inverse Hessian-vector product (iHVP), which is difficult to approximate efficiently. We introduce an algorithm (ASTRA) which uses the EKFAC-preconditioner on Neumann series iterations to arrive at an accurate iHVP approximation for TDA. ASTRA is easy to tune, requires fewer iterations than Neumann series iterations, and is more accurate than EKFAC-based approximations. Using ASTRA, we show that improving the accuracy of the iHVP approximation can significantly improve TDA performance.

📄 PDF Abstract BibTeX arXiv:2507.14740

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Better Hessians Matter: Studying the Impact of Curvature Approximations in Influence Functions

2025-09-27 · Steve Hong, Runa Eschenhagen, Bruno Mlodozeniec, Richard Turner arxiv

Influence functions offer a principled way to trace model predictions back to training data, but their use in deep learning is hampered by the need to invert a large, ill-conditioned Hessian matrix. Approximations such a…

Robust Attribution Regularization

2019-05-23 · NeurIPS 2019 12 · Jiefeng Chen, Xi Wu, Vaibhav Rastogi, YIngyu Liang 외

An emerging problem in trustworthy machine learning is to train models that produce robust interpretations for their predictions. We take a step towards solving this problem through the lens of axiomatic attribution of n…

CAWA: An Attention-Network for Credit Attribution

2019-11-26 · Saurav Manchanda, George Karypis

Credit attribution is the task of associating individual parts in a document with their most appropriate class labels. It is an important task with applications to information retrieval and text summarization. When label…

Information RetrievalMultilabel Text ClassificationRetrievalSentence+3

See Beyond a Single View: Multi-Attribution Learning Leads to Better Conversion Rate Prediction

2025-08-21 · Sishuo Chen, Zhangming Chan, Xiang-Rong Sheng, Lei Zhang 외 arxiv

Conversion rate (CVR) prediction is a core component of online advertising systems, where the attribution mechanisms-rules for allocating conversion credit across user touchpoints-fundamentally determine label generation…

Efficient Ensembles Improve Training Data Attribution

2024-05-27 · Junwei Deng, Ting-Wei Li, Shichang Zhang, Jiaqi Ma

Training data attribution (TDA) methods aim to quantify the influence of individual training data points on the model predictions, with broad applications in data-centric AI, such as mislabel detection, data selection, a…

AttributeComputational Efficiency