Machine Learning Reading Group

Department of Computer Science
Tufts University

Next meeting

Date Paper Discussion Leader
Wednesday 11/17, 9am The Kernel-Adatron Algorithm: a Fast and Simple Learning Procedure for Support Vector Machines ICML 1998 Gabriel Wachman
Wednesday 10/27, 9am Cyclic pattern kernels for predictive graph mining KDD04 Gabriel Wachman
Wednesday 10/20, 9am User Re-Authentication via Mouse Movements (paper circulated) Maja Pusara
Wednesday 10/6, 9am DISTILL: Learning Domain-Specific Planners by Example ICML03 Alan Carlin
Wednesday 9/29, 9am Learning to Map between Ontologieson the Semantic Web in WWW 02 Roni Khardon
Wednesday 9/22, 9am Approximate Policy Iteration with a Policy Language Bias in NIPS 03 Saket Joshi

Previous Papers (Spring/Summer 2004)

Paper Discussion Leader
Text Bundling: Statistics Based Data-Reduction Roni Khardon
Stochastic Local Search in k-term DNF Learning Marta Arias
Multi-Instance Learning Based Web-Mining Kai Jiang
Using the Triangle Inequality to Accelerate k-Means Wenting Zhou
A Geometric Framework for Unsupervised Anomaly Detection: Detecting Intrusions in Unlabeled Data Doug Stetson
Theory and Implenetation in Learning Logic Marta Arias
Mismatch String Kernels for SVM Protein Classification Peter Waltman
Inductive Policy Selection for First-Order Markov Decision Processes Roni Khardon
New Ranking Algorithms for Parsing and Tagging: Kernels over Discrete Structures, and the Voted Perceptron.
Michael Collins and Nigel Duffy. ACL 2002.
Gabriel Wachman
Efficient Algorithms for Learning to Play Repeated Games Against Computationally Bounded Adversaries.
Y. Freund, M. Kearns, Y. Mansour, D. Ron, R. Rubinfeld, and R. Schapire. FOCS 1995.
Jason Kroll
Bottom-Up Learning of Logic Programs for Information Extraction from Hypertext Documents (PKDD 2003) Roni Khardon

Paper Sources

Sources for Papers: Please note some of these (ECML, PKDD, ILP from the LNCS/LNAI site) require access from within the Tufts domain. If working from elsewhere you may need to ssh and then lynx to the sites. ICML is accessible from anywhere.

Concrete list to choose from so far only from ICML03 and PKDD03 (please feel free to suggest other papers):