BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Sample-Efficient Active Learning from Human Feedback
DTSTART:20260107T111500
DTEND:20260107T120000
DTSTAMP:20260407T182220Z
UID:58d5576bd4bbb7c707325d7124d054ab45e504681a19cca61bfcb0c1
CATEGORIES:Conferences - Seminars
DESCRIPTION:Belen Martin-Urcelay\, PhD Giorgia Tech\, USA\nAbstract\nAlign
 ing machine learning systems with human intent requires querying experts\,
  but collecting high-quality feedback at scale is often prohibitively expe
 nsive. This talk presents active learning frameworks designed to minimize 
 sample complexity by exploiting structure in the data and the feedback mec
 hanism.\nThe talk focuses on linear classifier learning within pre-trained
  embedding spaces. While traditional methods rely on simple labeling\, we 
 demonstrate that leveraging the embedding geometry allows for richer forms
  of feedback\, such as rankings. We analyze how these richer queries theor
 etically reduce sample complexity. We further develop active learning stra
 tegies that explicitly balance informational value against annotation cost
 s\, leading empirically to significant cost reductions.  The talk conclud
 es by pointing out future avenues for more efficient human-in-the-loop lea
 rning.\n\nBiography\nBelén Martín-Urcelay is a Ph.D. candidate at Georgi
 a Tech\, advised by Matthieu Bloch and Christopher Rozell. She earned h
 er B.Sc. and M.Sc. in Telecommunications Engineering from Universidad de N
 avarra before joining Georgia Tech for her doctorate. Her research focuses
  on active learning\, human feedback\, and reinforcement learning\, with a
 n emphasis on leveraging knowledgeable teachers (whether human experts or 
 powerful machine learning models) to enhance learning algorithm performanc
 e. She has collaborated internationally\, including a research visit to ET
 H Zurich with Andreas Krause. Outside of research\, she is passionate abo
 ut teaching\, mentoring students\, and promoting women in STEM fields.\n 
LOCATION:ME C2 405 https://plan.epfl.ch/?room==ME%20C2%20405
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR