SecureCut: Federated Gradient Boosting Decision Trees with Efficient Machine Unlearning
This addresses the need for efficient data deletion in federated learning systems for companies subject to privacy laws, representing a novel application in VFL scenarios.
The paper tackles the problem of enabling data removal, or machine unlearning, in Vertical Federated Learning (VFL) to comply with privacy regulations, proposing a Gradient Boosting Decision Tree framework that achieves superior model utility and forgetfulness compared to state-of-the-art methods without retraining from scratch.
In response to legislation mandating companies to honor the \textit{right to be forgotten} by erasing user data, it has become imperative to enable data removal in Vertical Federated Learning (VFL) where multiple parties provide private features for model training. In VFL, data removal, i.e., \textit{machine unlearning}, often requires removing specific features across all samples under privacy guarentee in federated learning. To address this challenge, we propose \methname, a novel Gradient Boosting Decision Tree (GBDT) framework that effectively enables both \textit{instance unlearning} and \textit{feature unlearning} without the need for retraining from scratch. Leveraging a robust GBDT structure, we enable effective data deletion while reducing degradation of model performance. Extensive experimental results on popular datasets demonstrate that our method achieves superior model utility and forgetfulness compared to \textit{state-of-the-art} methods. To our best knowledge, this is the first work that investigates machine unlearning in VFL scenarios.