Abstract
With e-commerce as the motivating example, we introduce a new multi-armed bandit setting where arms are grouped inside “ordered” categories. We conduct an analysis on real data to highlight that those ordered categories actually exist in practice. Finally, we provide algorithms that fully leverage the structure of the model and experimental results show the superiority of our proposed policies.
| Original language | English |
|---|---|
| Journal | CEUR Workshop Proceedings |
| Volume | 2621 |
| Publication status | Published - 1 Jan 2020 |
| Externally published | Yes |
| Event | 1st Joint Conference of the Information Retrieval Communities in Europe, CIRCLE 2020 - Samatan, Gers, France Duration: 6 Jul 2020 → 9 Jul 2020 |
Keywords
- E-commerce
- Multi-armed bandits
- Recommender systems
Fingerprint
Dive into the research topics of 'Categorized bandits'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver