Continual Learning and Private Unlearning
About
As intelligent agents become autonomous over longer periods of time, they may eventually become lifelong counterparts to specific people. If so, it may be common for a user to want the agent to master a task temporarily but later on to forget the task due to privacy concerns. However enabling an agent to \emph{forget privately} what the user specified without degrading the rest of the learned knowledge is a challenging problem. With the aim of addressing this challenge, this paper formalizes this continual learning and private unlearning (CLPU) problem. The paper further introduces a straightforward but exactly private solution, CLPU-DER++, as the first step towards solving the CLPU problem, along with a set of carefully designed benchmark problems to evaluate the effectiveness of the proposed solution. The code is available at https://github.com/Cranial-XIX/Continual-Learning-Private-Unlearning.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Knowledge Retention | MMLU (full) | MMLU Accuracy43.6 | 60 | |
| Machine Unlearning | TOFU Forget 1% | Aggregation Score50 | 54 | |
| Machine Unlearning | TOFU forget05 1.0 | Model Utility (MU)62 | 53 | |
| Machine Unlearning | TOFU 1.0 (forget01) | Average Score57 | 53 | |
| Classification | MLLMU-Bench Forget Set | Accuracy60.83 | 51 | |
| Machine Unlearning | TOFU | Forget Quality (FQ)0.0268 | 43 | |
| Multimodal Understanding | MMBench | Classification Score89.65 | 42 | |
| Visual Question Answering (VQA) | MLLMU-Bench 5% (forget) | Accuracy (Classification)49.17 | 42 | |
| Generation | MLLMU-Bench Forget Set | Rouge Score50.2 | 37 | |
| Multimodal Machine Unlearning Evaluation | MLLMU-Bench Forget Set | Classification Accuracy40.38 | 36 |