March 31, 2023 14:25
TrustML Young Scientist Seminar #63 2023028 Talk by Mimee Xu (New York University) thumbnails

Description

The 63rd Seminar
Date and Time: March 28th 10:00 am – 11:00 am(JST)

Speaker: Mimee Xu (New York University)
Title: Netflix and Forget
Short Abstract:
People break up, miscarry, and lose loved ones. Their online streaming and shopping recommendations, however, do not necessarily update, and may serve as unhappy reminders of their loss. When users want to renege on their past actions, they expect the recommender platforms to erase selective data at the model level. Ideally, given any specified user history, the recommender can unwind or “forget”, as if the record was not part of training. To that end, this paper focuses on simple but widely deployed bi-linear models for recommendations based on matrix completion. Without incurring the cost of re-training, and without degrading the model unnecessarily, we develop Unlearn-ALS by making a few key modifications to the fine-tuning procedure under Alternating Least Squares optimisation, thus applicable to any bi-linear models regardless of the training procedure. We show that Unlearn-ALS is consistent with retraining without \emph{any} model degradation and exhibits rapid convergence, making it suitable for a large class of existing recommenders.

Bio:
Mimee Xu is a PhD student in New York University advised by León Bottou, with a focus on AI Privacy. Her previous work explored secure and equitable evaluation of private training data for machine learning.