Debiasing Machine Unlearning with Counterfactual Examples

TL;DR

This paper addresses bias issues in machine unlearning for the right to be forgotten, proposing a causal, intervention-based method using counterfactual examples to improve unlearning effectiveness and fairness.

Contribution

It introduces a novel approach combining causal analysis and counterfactual examples to mitigate bias in machine unlearning processes.

Findings

Outperforms existing unlearning baselines on evaluation metrics

Effectively reduces data-level and algorithm-level biases

Maintains semantic data consistency during unlearning

Abstract

The right to be forgotten (RTBF) seeks to safeguard individuals from the enduring effects of their historical actions by implementing machine-learning techniques. These techniques facilitate the deletion of previously acquired knowledge without requiring extensive model retraining. However, they often overlook a critical issue: unlearning processes bias. This bias emerges from two main sources: (1) data-level bias, characterized by uneven data removal, and (2) algorithm-level bias, which leads to the contamination of the remaining dataset, thereby degrading model accuracy. In this work, we analyze the causal factors behind the unlearning process and mitigate biases at both data and algorithmic levels. Typically, we introduce an intervention-based approach, where knowledge to forget is erased with a debiased dataset. Besides, we guide the forgetting procedure by leveraging counterfactual examples, as they maintain semantic data consistency without hurting performance on the remaining dataset. Experimental results demonstrate that our method outperforms existing machine unlearning baselines on evaluation metrics.