FreqX: Analyze the Attribution Methods in Another Domain

TL;DR

FreqX is a novel interpretability method for personalized federated learning that combines signal processing and information theory, providing fast, detailed explanations of model attributions and concepts.

Contribution

Introduces FreqX, a new interpretability approach for PFL that is faster and more informative than existing methods, addressing privacy and fairness challenges.

Findings

FreqX explains both attribution and concept information.

FreqX is at least 10 times faster than baseline methods.

FreqX effectively enhances interpretability in PFL scenarios.

Abstract

Personalized Federal learning(PFL) allows clients to cooperatively train a personalized model without disclosing their private dataset. However, PFL suffers from Non-IID, heterogeneous devices, lack of fairness, and unclear contribution which urgently need the interpretability of deep learning model to overcome these challenges. These challenges proposed new demands for interpretability. Low cost, privacy, and detailed information. There is no current interpretability method satisfying them. In this paper, we propose a novel interpretability method \emph{FreqX} by introducing Signal Processing and Information Theory. Our experiments show that the explanation results of FreqX contain both attribution information and concept information. FreqX runs at least 10 times faster than the baselines which contain concept information.