In this Few-Shot Class-Incremental Learning (FSCIL) scenario, existing methods either introduce extra learnable components or rely on a frozen feature extractor to mitigate catastrophic forgetting and overfitting problems.

Then, we propose a series of DP-SignRP algorithms that leverage the robustness of the "sign flipping probability" of random projections.

We used reliable public data from 1970 to 2020 to curate attributes like prior criminal counts and recidivism outcomes. The dataset contains large number of samples from five racial groups, in addition to information like sex and age (at judgment and first offense).

For example, in healthcare domains, such annotations may be obtained by asking clinicians what they think would happen to the patient if a different treatment were to be used.