Machine Learning Ph.D. Seminar

Fall 2012 Schedule

September 11th - Borja Balles, Universitat Politecnica de Catalunya (UPC)
Learning Markovian Models from Time-Evolving Data Streams
Abstract
September 18th - Maria Pershina , New York University
Title: Finite Automata as a Supermodel for Sequence to Sequence Mappings
Abstract
September 25th - Andres Munoz, New York University
Title: New Theory and Algorithm for Learning with Drifting Distributions
Abstract
October 16 - University Holiday
Friday October 19 - New York Machine Learning Symposium
October 23th - Mehryar Mohri , New York University
Title: Accuracy at the top
Abstract
October 30 - Algorithmic Learning Theory Conference (No talks scheduled).
November 6th - Sham Kakade, Microsoft Research
Title: Tensor Decompositions for Learning Latent Variable Models
Abstract
November 13th - Kareem Amin, UPenn
Title:Budget optimization for sponsored search: Censored Learning in MDPs
Abstract
November 20th - John Langford, Microsoft Research
Title: Bidding bandits
Abstract
This talk has been rescheduled to next semester
November 27th - Jennifer Wortman, UCLA, Microsoft Research
Title: Efficient Market Making via Convex Optimization and a Connection to Online Learning.
Abstract
December 4 - NIPS (No seminar)
December 11 - Gunnar Raetsch Memorial Sloan Kettering Cancer Center
Title: TBD

Spring 2012 Schedule

February 21st - Panos Ipeirotis, Stern NYU
Towards a theory model for product search.
Abstract
Feb 28th - Umar Syed, Google Research
Graphical models for bandit problems
Abstract
March 6th - Vladimir Vapnik, NEC Labs
Learning using priviledged information
Abstract
March 20th - Yann LeCun , CIMS
Learning Hierarchies of invariant features
Abstract
April 3rd - Steven Hanneke , Carnegie Mellon
The label complexity of active learning
Abstract
April 10th - Slav Petrov , Google Research

Details