We present a machine unlearning approach that is both retraining- and label-free. Most existing machine unlearning approaches require a model to be fine-tuned to remove information while preserving performance. This is computationally expensive and necessitates the storage of the whole dataset for the lifetime of the model. Retraining-free approaches often utilise Fisher information, which is derived from the loss and requires labelled data which may not be available. Thus, we present an extension to the Selective Synaptic Dampening algorithm, substituting the diagonal of the Fisher information matrix for the gradient of the l2 norm of the model output to approximate sensitivity. We evaluate our method in a range of experiments using ResNet18 and Vision Transformer. Results show our label-free method is competitive with existing state-of-the-art approaches.

翻译：我们提出一种既无需重新训练、也无需标签的机器遗忘方法。现有大多数机器遗忘方法需要通过微调模型来移除信息同时保持性能，这不仅计算成本高昂，还需在模型整个生命周期内存储全部数据集。免重训方法通常利用源自损失函数的Fisher信息，但这需要标签数据，而此类数据可能无法获取。为此，我们扩展了选择性突触抑制算法，用模型输出的L2范数梯度替代Fisher信息矩阵的对角线，以近似估计敏感度。我们在使用ResNet18与Vision Transformer的系列实验中评估了该方法。结果表明，我们的无标签方法可与现有最优方法相媲美。