The Mirrored Influence Hypothesis: Efficient Data Influence Estimation by Harnessing Forward Passes

arXiv:2402.08922v1 Announce Type: new
Abstract: Large-scale black-box models have become ubiquitous across numerous applications. Understanding the influence of individual training data sources on predictions made by these models is crucial for improving their trustworthiness. Current influence estimation techniques involve computing gradients for every training point or repeated training on different subsets. These approaches face obvious computational challenges when scaled up to large datasets and models.
In this paper, we introduce and explore the Mirrored Influence Hypothesis, highlighting a reciprocal nature …

abstract applications arxiv become box computing cs.lg current data data sources every hypothesis influence predictions scale stat.ml training training data type understanding