Arpan Mukherjee
Scholar

Arpan Mukherjee

Google Scholar ID: jAS9pzQAAAAJ
Imperial College London
Sequential experimental designBest Arm IdentificationRLHF
Citations & Impact
All-time
Citations
88
 
H-index
6
 
i10-index
3
 
Publications
18
 
Co-authors
9
list available
Publications
18 items
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
  • Publications: - Group testing for combinatorial bandits, accepted in TMLR; - Preference-centric bandits, available on arXiv; - Risk-sensitive bandits, accepted to AISTATS 2025.
Research Experience
  • Current position: Informed-AI postdoctoral researcher at the IPC lab in Imperial College London, working on the theory of language models with Prof. Deniz Gündüz.
Education
  • Ph.D.: Department of Electrical, Computer and Systems Engineering (ECSE), Rensselaer Polytechnic Institute (RPI), Advisor: Prof. Ali Tajer; M.Sc./B.Sc.: Indian Institute of Technology, Kharagpur, Advisor: Prof. Mrityunjoy Chakraborty.
Background
  • Research Interests: Problems at the intersection of signal processing, statistics, and machine learning; main research areas include sequential experimental design, multi-armed bandits, optimal stopping, active learning, data-efficient decision making, identification problems, robustness, and risk-sensitivity. Recently, started looking into reinforcement learning from human feedback (RLHF), focusing on sample complexity, preference diversity, and multi-objective RLHF.
Miscellany
  • Personal website built using Jekyll with the al-folio theme, hosted by GitHub Pages. Photos from Unsplash.