Interference-Aware Multi-Task Unlearning

干扰感知多任务机器遗忘

Abstract: Machine unlearning aims to remove the contribution of designated training data from a trained model while preserving performance on the remaining data. Existing work mainly focuses on single-task settings, whereas modern models often operate in multi-task setups with shared backbones, where removing supervision for one task or instance can unintentionally affect others.

摘要： 机器遗忘旨在从已训练的模型中移除指定训练数据的贡献，同时保持模型在剩余数据上的性能。现有的研究主要集中在单任务设置上，而现代模型通常在具有共享主干的多任务架构下运行，在这种架构中，移除某个任务或实例的监督信息可能会无意中影响其他任务。

We introduce multi-task unlearning with two settings: full-task unlearning, which removes a target instance from all tasks, and partial-task unlearning, which removes supervision only from selected tasks. We show that shared parameters couple the forget and retain sets, causing task-level interference on non-target tasks and instance-level interference on other instances.

我们引入了两种多任务遗忘设置：全任务遗忘（full-task unlearning），即从所有任务中移除目标实例；以及部分任务遗忘（partial-task unlearning），即仅从选定的任务中移除监督信息。我们指出，共享参数将“遗忘集”与“保留集”耦合在一起，导致了非目标任务上的任务级干扰以及其他实例上的实例级干扰。

To address this issue, we propose an interference-aware framework that combines task-aware gradient projection, which constrains updates within task-specific subspaces, with instance-level gradient orthogonalization, which reduces conflicts between forget and retain signals.

为了解决这一问题，我们提出了一个干扰感知框架，该框架结合了任务感知梯度投影（将更新限制在特定任务的子空间内）和实例级梯度正交化（减少遗忘信号与保留信号之间的冲突）。

Experiments on two multi-task computer vision benchmarks across five tasks show that our method achieves effective unlearning while maintaining strong generalization, reducing UIS compared with the strongest baseline by 30.3% in full-task unlearning and 52.9% in partial-task unlearning.

在五个任务的两个多任务计算机视觉基准测试上的实验表明，我们的方法在实现有效遗忘的同时保持了强大的泛化能力；与最强的基准方法相比，全任务遗忘下的 UIS 指标降低了 30.3%，部分任务遗忘下的 UIS 指标降低了 52.9%。