PulseAugur
实时 13:42:25

New dual representation for influence functions improves efficiency

Researchers have developed a new dual representation for influence functions, which can efficiently estimate changes in model parameters and outputs. This method scales with dataset size rather than model size, offering an advantage for large models where traditional influence function evaluation is infeasible. However, the approach is currently limited to linearizable models and requires substantial matrix materialization. AI

影响 Introduces a more efficient method for analyzing model behavior, potentially aiding in debugging and understanding large-scale machine learning models.

排序理由 The cluster contains an academic paper detailing a new research method in machine learning.

在 arXiv stat.ML 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

New dual representation for influence functions improves efficiency

报道来源 [2]

  1. arXiv stat.ML TIER_1 English(EN) · Zhenhuan Sun, Shahrokh Valaee ·

    Extending Kernel Trick to Influence Functions

    arXiv:2605.11239v1 Announce Type: cross Abstract: In this paper, we present a dual representation of the influence functions, whose computational complexity scales with dataset size rather than model size. Both analytically and experimentally, we show that this representation can…

  2. arXiv stat.ML TIER_1 English(EN) · Shahrokh Valaee ·

    Extending Kernel Trick to Influence Functions

    In this paper, we present a dual representation of the influence functions, whose computational complexity scales with dataset size rather than model size. Both analytically and experimentally, we show that this representation can be an efficient alternative to the original influ…