Home
Active learning has been a topic of significant research over the past several decades with much attention devoted to both theoretical and practical considerations. A variety of algorithms and sampling paradigms have been proposed and studied, but roughly speaking, this line of research focuses on how to make feedback driven decisions about data collection, and how to leverage this power for efficient learning. Research in this area stems from a range of communities including signal processing, machine learning, statistics, and information theory, including both theoreticians and practitioners. One aim of this workshop is to bring this diverse collection of researchers together.
Despite attention from both theoreticians and practitioners, there remains a glaring disconnect between the two lines of research, and the other aim of this workshop is to find concrete directions toward bridging this divide. Many of the algorithms with strong statistical guarantees either make strong modeling assumptions or suffer from computational inefficiencies while many algorithms achieving empirical successes are less amenable to theoretical analysis. By bringing both theoreticians and practitioners together, we hope to identify future research directions that address this disconnect.
The workshop will be part of ICML 2015 in Lille, France.
Instructions for contributed submissions are here.
Important Dates
Deadline for submission of papers: May 1st, 2015
Notification of acceptance: May 10th, 2015
Workshop: July 10th, 2015
Invited Speakers
Adam Kalai (Microsoft Research)
Andreas Krause (ETH Zurich)
John Langford (Microsoft Research)
Maja Temerinac-Ott (Univ. of Freiburg)
Jeff Schneider (Carnegie Mellon University)
Burr Settles (Duolingo, FAWM)
Organizers
Akshay Krishnamurthy (Carnegie Mellon University)
Aaditya Ramdas (Carnegie Mellon University)
Nina Balcan (Carnegie Mellon University)
Aarti Singh(Carnegie Mellon University)