arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.22124cs.LGcs.AI
收藏

高效实用型机器去学习与隐式梯度手术

Efficient Utility-Preserving Machine Unlearning with Implicit Gradient Surgery

Institute of Artificial Intelligence, Beihang University(北航人工智能研究院);Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing, Beihang University(北京未来区块链与隐私计算先进创新中心);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);University of Amsterdam(阿姆斯特丹大学);Tsinghua University(清华大学);YanTron Technology Co.Ltd(燕tron技术有限公司)

Shiji Zhou, Tianbai Yu, Zhi Zhang, Heng Chang, Xiao Zhou, Dong Wu, Han Zhao

AI总结:

本文提出了一种高效实用型机器去学习方法,通过隐式梯度手术实现去学习与实用性保持的平衡。

摘要:

机器去学习(MU)旨在高效地从预训练模型中移除敏感或有害的记忆。关键挑战是平衡去学习效果与实用性保持之间的潜在权衡,这涉及在定义中遗忘不良信息的同时保持模型的原始性能。一种可能的解决方法是使用多目标优化,共同优化去学习和实用性保持的目标。然而,现有多目标方法只能保证找到帕累托最优解,而没有精细控制,导致去学习目标的欠优化。为此,我们首先将MU建模为一个约束优化问题,即在实用性损失的有界增加约束下优化去学习目标。然后我们证明了解决这个问题优化问题等同于对去学习目标进行单方面梯度手术。为了解决梯度手术带来的额外计算成本,我们提出了一种隐式梯度手术方法,仅通过一次反向传播近似解决上述约束优化问题,从而实现高效的实用性保持MU。理论上,我们提供了算法的紧收敛分析。实验上,我们的广泛实验表明,所提出的方法在现有基线中取得了更好的权衡结果。代码可在https://github.com/anseryuer/EUPMU-Efficient-Utility-Preserving-Machine-Unlearning获取。

英文摘要:

Machine unlearning (MU) aims to efficiently remove sensitive or harmful memory from a pre-trained model. The key challenge is to balance the potential tradeoff between unlearning efficacy and utility preservation, which involves forgetting undesirable information as defined while maintaining the model's original performance. One potential way to tackle this problem is to use multi-objective optimization to jointly optimize both the unlearning and utility preservation objectives. However, existing multi-objective methods only guarantee finding a Pareto-optimal solution without fine-grained control, which causes under-optimization of the unlearning objective. To this end, we first model MU as a constrained optimization problem, that is, optimizing the unlearning objective under the constraint of a bounded increase for utility loss. We then show that solving this optimization problem is equivalent to unilateral gradient surgery on the unlearning objective. To resolve the additional computational cost brought by gradient surgery, we propose an implicit gradient surgery method, which approximates the solution to the aforementioned constrained optimization problem via only one backpropagation, thereby achieving efficient utility-preserving MU. Theoretically, we provide a tight convergence analysis of the algorithm. Empirically, our extensive experiments show that the proposed algorithm achieves better tradeoff results than existing baselines. Codes are available at https://github.com/anseryuer/EUPMU-Efficient-Utility-Preserving-Machine-Unlearning.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑