Scholar
Pierre Clavier
Google Scholar ID: -KnIaGsAAAAJ
Research Scientist at Cohere
Citations & Impact
Citations
247
H-index
6
i10-index
4
Publications
12
Co-authors
14
list available
Publications
Resume (English only)
Academic Achievements
- 2025: Paper "ShiQ" accepted to NeurIPS 2025.
- 2024: Two papers accepted to NeurIPS 2024 — TC-MDP (a new algorithm for Robust RL) and a theoretical paper on optimal sample complexity of Robust MDPs (co-authored with Laixi).
- 2024: One paper accepted at ICML 2024 on Bandits with Variational Inference; one oral presentation at UAI 2024 on Robust MDP theory.
- May 2024: Released three preprints — RRLS (a new benchmark for Robust RL), and two new algorithms: ExpectRL and TC-MDP.
- January 2025: Contributed to Cohere’s AI model "Command A", fine-tuned using theoretically grounded RLHF algorithms CoPG and SRPO.
- November 2024: Successfully defended PhD thesis.
