Efficient Thompson Sampling for Online Matrix-Factorization Recommendation

By Jaya Kawale, Hung Hai Bui, Branislav Kveton, Long Tran-Thanh, Sanjay Chawla. 2015

196 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

DRN: A Deep Reinforcement Learning Framework for News Recommendation (2018, background): “Recently, some people try to combine bandit with clustering based collaborative filtering [14], and matrix factorization [6, 21, 32, 42, 43, 51], in order to model more complex user and item relationship, and utilize the social network relationship in determining the reward function.”
Learning Algorithms for Active Learning (2017, background): “…systems, for example, ratings data for existing users can inform a strategy that efﬁciently elicits preferences for new users who lack prior rating data, thus boot-strapping the system quickly out of the cold-start setting (Golbandi et al., 2010; 2011; Sun et al., 2013a; Kawale et al., 2015).”
A Tutorial on Thompson Sampling (2017, methodology): “…revenue management [7], marketing [8], Monte Carlo tree search [9], A/B testing [10], Internet advertising [10, 11, 12], recommendation systems [13], hyperparameter tuning [14], and arcade games [15]; and have been used at several companies, including Microsoft [10], Google [6, 16], LinkedIn…”
Bandit Algorithms (2020, background): “The Kiefer–Wolfowitz theorem is due to Kiefer and Wolfowitz [1960]. The algorithm in Note 3 is due to Fedorov [1972].”
Learning Hidden Features for Contextual Bandits (2016, methodology): “For example, the performance of PTS and hLinUCB boosted on the first testing day.”
Meta-Adaptive Nonlinear Control: Theory and Algorithms (2021, methodology): “Outside the context of control, such bilinear models are also studied in representation learning [4, 5] and factorization bandits [29, 30].”
Towards Conversational Recommender Systems (2016, background): “In [13], the authors extend Thompson Sampling for PMF with a Kalman ﬁlter to track changing preferences over time.”
Factorization Bandits for Interactive Recommendation (2017, methodology): “, the user-item matrix is one rank one and all the latent factors can be properly discretized beforehand (Kawale et al. 2015)).”