Ticketed Learning-Unlearning Schemes
Ghazi, Badih, Kamath, Pritish, Kumar, Ravi, Manurangsi, Pasin, Sekhari, Ayush, Zhang, Chiyuan
–arXiv.org Artificial Intelligence
We consider the learning--unlearning paradigm defined as follows. First given a dataset, the goal is to learn a good predictor, such as one minimizing a certain loss. Subsequently, given any subset of examples that wish to be unlearnt, the goal is to learn, without the knowledge of the original training dataset, a good predictor that is identical to the predictor that would have been produced when learning from scratch on the surviving examples. We propose a new ticketed model for learning--unlearning wherein the learning algorithm can send back additional information in the form of a small-sized (encrypted) ``ticket'' to each participating training example, in addition to retaining a small amount of ``central'' information for later. Subsequently, the examples that wish to be unlearnt present their tickets to the unlearning algorithm, which additionally uses the central information to return a new predictor. We provide space-efficient ticketed learning--unlearning schemes for a broad family of concept classes, including thresholds, parities, intersection-closed classes, among others. En route, we introduce the count-to-zero problem, where during unlearning, the goal is to simply know if there are any examples that survived. We give a ticketed learning--unlearning scheme for this problem that relies on the construction of Sperner families with certain properties, which might be of independent interest.
arXiv.org Artificial Intelligence
Jun-27-2023
- Country:
- Asia > Thailand (0.04)
- South America > Chile
- North America > United States
- Massachusetts > Middlesex County > Cambridge (0.04)
- Europe > United Kingdom
- England > Cambridgeshire > Cambridge (0.04)
- Genre:
- Research Report (0.81)
- Industry:
- Information Technology > Security & Privacy (0.46)
- Technology: