A Peep into the Future: Adversarial Future Encoding in Recommendation

Abstract

Personalized recommendation often relies on user historical behaviors to provide items for users. It is intuitive that future information also contains essential messages as supplements to user historical behaviors. However, we cannot directly encode future information into models, since we are unable to get future information in online serving. In this work, we propose a novel adversarial future encoding (AFE) framework to make full use of informative future features in different types of recommendation models. Specifically, AFE contains a future-aware discriminator and a generator. The future-aware discriminator takes both common features and future features as inputs, working as a recommendation prophet to judge user-item pairs. In contrast, the generator is considered as a challenger, which generates items with only common features, aiming to confuse the future-aware prophet. The future-aware discriminator can inspire the generator (to be deployed online) to produce better results. We further conduct a multi-factor optimization to enable a fast and stable model convergence via the direct learning and knowledge distillation losses. Moreover, we have adopted AFE on both a list-wise RL-based ranking model and a point-wise ranking model to verify its universality. In experiments, we conduct sufficient evaluations on two large-scale datasets, achieving significant improvements on both offline and online evaluations. Currently, we have deployed AFE on a real-world system, affecting millions of users. The source code is in https://github.com/modriczhang/AFE.

Workshop on Learning and Evaluating Recommendations with Impressions (LERI)

“…Some works have already tried to use impressions to build better recommendation models in various ways: [4,5,16,33,38,39] use impression data to compute features, re-ranking, sampling and to learn biases. Furthermore [6,10,14,17,18,31,36] apply neural or deep-learning models including impressions. Most of these papers have been published in the last two years in conferences such as SIGIR, KDD, WWW and RecSys.…”