ICML2009 Awards Nominees

The ICML 2009 Awards Committee is pleased to announce the winners for this year's best paper awards. (Papers not marked as winners were honorable mentions.)

Best overall paper

A paper the committee feels contains innovative and creative results and best conveys deeper insights that are of interest to researchers both inside and outside of machine learning.

Archipelago: Nonparametric Bayesian Semi-Supervised Learning. Ryan Adams and Zoubin Ghahramani
paper ID: 258
(* Also a best student paper honorable mention)

Winner: Structure Preserving Embedding. Blake Shaw and Tony Jebara
paper ID: 418
(* Also a best student paper winner)

Optimal Reverse Prediction: A Unified Perspective on Supervised, Unsupervised and Semi-supervised Learning. Linli Xu, Martha White and Dale Schuurmans
paper ID: 578

Best application paper

A paper the committee feels best demonstrates the promise of machine learning's impact on applications.

Winner: Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations. Honglak Lee, Roger Grosse, Rajesh Ranganath and Andrew Ng
paper ID: 571
(* Also a best student paper winner)

Best student paper

Papers the committee judges to be a top-notch machine-learning paper and whose first author is a graduate student. We are grateful to a generous contribution from Springer, the publishers of "Machine Learning", for helping to fund this award.

Tractable Nonparametric Bayesian Inference in Poisson Processes with Gaussian Process Intensities. Ryan Adams, Iain Murray and David MacKay
paper ID: 255

Large Margin Training for Hidden Markov Models with Partially Observed States. Trinh-Minh-Tri Do and Thierry Artieres
paper ID: 298

A Scalable Framework for Discovering Coherent Co-clusters in Noisy Data. Meghana Deodhar, Joydeep Ghosh, Gunjan Gupta, Hyuk Cho and Inderjit Dhillon
paper ID: 316

Winner: BoltzRank: Learning to Maximize Expected Ranking Gain. Maksims Volkovs and Richard Zemel
paper ID: 498

Winner: Blockwise Coordinate Descent Procedures for the Multi-task Lasso, with Applications to Neural Semantic Basis Discovery. Han Liu, Mark Palatucci and Jian Zhang
paper ID: 168
(* Also a best overall paper honorable mention)

Best 10-year paper

A paper published in ICML 1999 the committee feels has had the most significant and lasting impact.

Winner: Transductive Inference for Text Classification using Support Vector Machines. Thorsten Joachims

Honorable mentions:

Policy Invariance Under Reward Transformations: Theory and Application to Reward Shaping. Andrew Y. Ng, Daishi Harada, Stuart J. Russell

Least-Squares Temporal Difference Learning. Justin A. Boyan

All award winning papers received certificates and a check for $800. All honorable mentions received certificates and a check for $200. Awards were presented in a special session at the conference. The best 10-year paper winner gave a "retrospective" talk during the special session.

ICML 2009 Awards Committee

Francis Bach
Leon Bottou
Andrea Danyluk
Yann LeCun
Michael Littman
Michele Sebag
Stefan Wrobel