Pfeed: Generating near real-time personalized feeds using precomputed embedding similarities

arXiv:2402.16073v1 Announce Type: cross
Abstract: In personalized recommender systems, embeddings are often used to encode customer actions and items, and retrieval is then performed in the embedding space using approximate nearest neighbor search. However, this approach can lead to two challenges: 1) user embeddings can restrict the diversity of interests captured and 2) the need to keep them up-to-date requires an expensive, real-time infrastructure. In this paper, we propose a method that overcomes these challenges in a practical, industrial setting. …

abstract approximate nearest neighbor arxiv challenges cs.ai cs.ir cs.lg customer diversity embedding embeddings encode near personalized real-time recommender systems retrieval search space systems type