Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note2 min read

FedXDS: Leveraging Model Attribution Methods to counteract Data Heterogeneity in Federated Learning

Source

arXiv

arxiv.org/abs/2606.31742v1

Read the source

Source

Plain-Language Brief

The paper introduces FedXDS, a method that uses explainable AI (XAI) to help mitigate performance degradation in federated learning caused by statistical heterogeneity between client data distributions. The approach involves identifying relevant features across clients through feature attribution techniques and selectively sharing these data elements for better model training. To protect sensitive information, the method employs metric privacy techniques that provide formal privacy guarantees.

What the Document Says

  • Explainable AI methods have been successful in identifying relevant features that drive deep learning model decisions.
  • The potential of XAI beyond providing model transparency has not been fully explored in adjacent machine learning domains.
  • The paper presents FedXDS, an approach to utilize XAI in federated learning to mitigate performance degradation caused by statistical heterogeneity between client data distributions.
  • FedXDS identifies task-relevant features through a single backward pass and enables selective data sharing that aligns client contributions.
  • To protect sensitive information, FedXDS incorporates metric privacy techniques for formal privacy guarantees.
  • Experimental results show higher accuracy and faster convergence of FedXDS compared to existing methods across varying client numbers and heterogeneity settings.
  • The paper provides theoretical privacy guarantees and empirically demonstrates robustness against membership inference and feature inversion attacks.
  • Code for FedXDS is available on GitHub.

Key Concepts Mentioned

  • Explainable AI
  • Federated learning
  • Statistical heterogeneity
  • Feature attribution
  • Selective data sharing
  • Metric privacy

Claims Made by the Document

  • XAI can be utilized in federated learning.
  • FedXDS can mitigate performance degradation caused by statistical heterogeneity between client data distributions.
  • Experimental results demonstrate higher accuracy and faster convergence of FedXDS compared to existing methods.

Evidence Used

  • The paper demonstrates the utilization of XAI in federated learning through the introduction of FedXDS.
  • Results from experiments are provided showing the effectiveness of FedXDS in various scenarios.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Explainable AI (XAI) methods have demonstrated significant success in recent years at identifying relevant features in input data that drive deep learning model decisions, enhancing interpretability for users. However, the potential of XAI beyond providing model transparency has remained largely unexplored in adjacent machine learning domains. In this paper, we show for the first time how XAI can be utilized in the context of federated learning. Specifically, while federated learning enables collaborative model training without raw data sharing, it suffers from performance degradation when client data distributions exhibit statistical heterogeneity. We introduce FedXDS (Federated Learning via XAI-guided Data Sharing), the first approach to utilize feature attribution techniques to identify precisely which data elements should be selectively shared between clients to mitigate heterogeneity. By employing propagation-based attribution, our method identifies task-relevant features through a single backward pass, enabling selective data sharing that aligns client contributions. To protect sensitive information, we incorporate metric privacy techniques that provide formal privacy guarantees while preserving utility. Experimental results demonstrate that our approach consistently achieves higher accuracy and faster convergence compared to existing methods across varying client numbers and heterogeneity settings. We provide theoretical privacy guarantees and empirically demonstrate robustness against both membership inference and feature inversion attacks. Code is available at https://github.com/MaxH1996/FedXDS.