Motti Goldberger

prof_pic_square.jpg

I am a Ph.D. candidate in Operations at the Yale School of Management, advised by Nils Rudi. Before Yale, I received a B.S. in Mathematics from the University of Maryland.

My research focuses on sequential learning and adaptive data collection, with an emphasis on developing algorithms for efficiently collecting information for ranking and selection. Much of my work is motivated by settings involving human evaluators, who often provide feedback in the form of pairwise comparisons or ratings. Applications of my research include LLM evaluation, hiring and admissions, clinical trials, simulation optimization, and tournament design.

Research

  1. ICML
    Optimal Top-k Identification from Pairwise Comparisons
    Motti Goldberger and Nils Rudi
    International Conference on Machine Learning (ICML), 2026
    arXiv Bib PMLR
  2. Fundamental Limitations of Fixed-Budget Best-Arm Identification
    Motti Goldberger
    Under review, 2026
    arXiv Bib
  3. Selecting the Best Few with Human Evaluators: Adaptively Combining Ratings and Pairwise Comparisons
    Motti Goldberger and Nils Rudi
    Under review, 2026
    arXiv Bib
  4. A Simple and General Approach to Top-k Selection
    Motti Goldberger and Nils Rudi
    Under review, 2026
    arXiv Bib
  5. Active Ranking via Minimizing the Sum of Top-k Regrets
    Motti Goldberger and Nils Rudi
    Work in progress, 2026
    An early version was accepted at the NeurIPS 2026 MLxOR Workshop.