Explore, exploit, and explain

James O. McInerney, Benjamin Lacker, Samantha Hansen, Karl Higley, Hugues Bouchard, Alois Gruson, Rishabh Mehrotra

2018 · 148 citations · 18 references

Concepts

TL;DR

The multi‑armed bandit framework balances exploration and exploitation in recommendation, and recent work has highlighted the need for exploration in new users, items, and non‑stationary settings, while also recognizing that explaining recommendations is essential for user understanding. We present the first principled method that jointly incorporates explanations and bandit decision‑making. Our approach simultaneously learns which explanations resonate with each user, identifies the optimal content for each user, and balances exploration with exploitation to handle uncertainty. Experiments on historical logs and live traffic in a large‑scale music recommendation service demonstrate a significant increase in user engagement.

Abstract

The multi-armed bandit is an important framework for balancing exploration with exploitation in recommendation. Exploitation recommends content (e.g., products, movies, music playlists) with the highest predicted user engagement and has traditionally been the focus of recommender systems. Exploration recommends content with uncertain predicted user engagement for the purpose of gathering more information. The importance of exploration has been recognized in recent years, particularly in settings with new users, new items, non-stationary preferences and attributes. In parallel, explaining recommendations ("recsplanations") is crucial if users are to understand their recommendations. Existing work has looked at bandits and explanations independently. We provide the first method that combines both in a principled manner. In particular, our method is able to jointly (1) learn which explanations each user responds to; (2) learn the best content to recommend for each user; and (3) balance exploration with exploitation to deal with uncertainty. Experiments with historical log data and tests with live production traffic in a large-scale music recommendation service show a significant improvement in user engagement.

References

18