Bandit Algorithms for Online Matrix Factorisation
Event details
| Date | 30.08.2023 |
| Hour | 10:00 › 12:00 |
| Speaker | Oscar Villemaud |
| Location | |
| Category | Conferences - Seminars |
EDIC candidacy exam
Exam president: Prof. Nicolas Flammarion
Thesis advisor: Prof. Matthias Grossglauser
Co-examiner: Prof. Olivier Lévêque
Abstract
Matrix factorisation is widely used for recommender systems, but it is typically used in a offline manner, which does not leverage the information we gain when an item is recommended. On the other hand, linear bandit algorithms update recommendations with every new data point, but they usually assume the exact knowledge of one of the two factors. In this work, we study how to perform matrix factorisation in an online manner without knowing any of the factors.
Background papers
- Ruslan Salakhutdinov and Andriy Mnih. Bayesian probabilistic matrix factorization using markov chain monte carlo. ICML, 2008
- Abbasi-Yadkori, Y., Pál, D., & Szepesvári, C., Improved algorithms for linear stochastic bandits, NeurIPS, 2011
- Jaya Kawale, Hung H Bui, Branislav Kveton, Long Tran-Thanh, and Sanjay Chawla. Efficient thompson sampling for online matrix-factorization recommendation, NIPS, 2015
Exam president: Prof. Nicolas Flammarion
Thesis advisor: Prof. Matthias Grossglauser
Co-examiner: Prof. Olivier Lévêque
Abstract
Matrix factorisation is widely used for recommender systems, but it is typically used in a offline manner, which does not leverage the information we gain when an item is recommended. On the other hand, linear bandit algorithms update recommendations with every new data point, but they usually assume the exact knowledge of one of the two factors. In this work, we study how to perform matrix factorisation in an online manner without knowing any of the factors.
Background papers
- Ruslan Salakhutdinov and Andriy Mnih. Bayesian probabilistic matrix factorization using markov chain monte carlo. ICML, 2008
- Abbasi-Yadkori, Y., Pál, D., & Szepesvári, C., Improved algorithms for linear stochastic bandits, NeurIPS, 2011
- Jaya Kawale, Hung H Bui, Branislav Kveton, Long Tran-Thanh, and Sanjay Chawla. Efficient thompson sampling for online matrix-factorization recommendation, NIPS, 2015
Practical information
- General public
- Free