Future Gradient Descent for Adapting the Temporal Shifting Data Distribution in Online Recommendation SystemDownload PDF

Published: 20 May 2022, Last Modified: 05 May 2023UAI 2022 PosterReaders: Everyone
Keywords: Recommendation systems, data distribution shift, meta learning, Incremental Update
TL;DR: We propose a future gradient descent algorithm that forecast the gradient information of future domain to overcome the data distribution shift issue in recommendation systems.
Abstract: One of the key challenges of learning an online recommendation model is the temporal domain shift, which causes the mismatch between the training and testing data distribution and hence domain generalization error. To overcome, we propose to learn a meta future gradient generator that forecasts the gradient information of the future data distribution for training so that the recommendation model can be trained as if we were able to look ahead at the future of its deployment. Compared with Batch Update, a widely used paradigm, our theory suggests that the proposed algorithm achieves smaller temporal domain generalization error measured by a gradient variation term in a local regret. We demonstrate the empirical advantage by comparing with various representative baselines.
Supplementary Material: zip
5 Replies