Advanced Machine Learning


* Home

* Handouts

* News

* Staff

* Papers

* Tutorials


* ML People


Read the following papers before each class.

Topic 1: Gaussians, Linear Models and PCA
Gaussian/PCA Face Recognition by Moghaddam, Jebara & Pentland
PCA on Natural Images by Rao & Ballard

Topic 2: Nonlinear Manifold Learning
Locally Linear Embedding by Saul & Roweis
Kernel PCA by Scholkopf, Smola & Muller
Semidefinite Embedding by Weinberger, Sha & Saul
Minimum Volume Embedding by Shaw & Jebara

Topic 3: Maximum Entropy, Discrimination and SVMs
A Maximum Entropy Approach to Natural Language Processing by Berger, Della Pietra & Della Pietra
Multitask Sparsity via Maximum Entropy Discrimination by Jebara (only need to read until pages 75-83 for now)

Topic 4: Logistic Regression and Conditional Random Fields
Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data by Lafferty, McCallum & Pereira
Majorization for CRFs by Jebara & Choromanska

Topic 5: Graphical Models and Structured Output Prediction
Review Chapters 11, 16 and 17 of Jordan
Cutting-Plane Training of Structural SVMs by Joachims, Finley & Yu
Structured Prediction with Relative Margin by Shivaswamy & Jebara

Topic 6: Beyond Junction Tree: High Tree-Width Models
Loopy Belief Propagation for Bipartite Maximum Weight b-Matching by Huang and Jebara
Perfect Graphs and Graphical Modeling by Jebara

Topic 7: Kernels and Structured Input Spacess
Exploiting generative models in discriminative classifiers by Jaakkola and Haussler
Probability Product Kernels by Jebara, Kondor and Howard

Topic 8: Feature and Kernel Selection
Feature Selection for SVMs by Weston, Mukherjee, Chapelle, Pontil, Poggio and Vapnik
Learning the Kernel for SVMs by Lanckriet et al.

Topic 9: Multi-Task Learning
Multitask Sparsity via Maximum Entropy Discrimination by Jebara (read rest of it)

Topic 10: Semi-Supervised Learning
Transductive Inference using SVMs by Joachims
Text Classification from Labeled and Unlabeled Documents using EM by K. Nigam, A. McCallum, S. Thrun and T. Mitchell
Semi-Supervised Learning Using Gaussian Fields and Harmonic Functions by Zhu, Ghahramani and Lafferty
Local and Global Consistency by Zhou et al.
Graph Construction and b-Matching for Semi-Supervised Learning by Jebara, Wang and Chang

Topic 11: Clustering, Graphs, Spectra and Matching
Spectral Clustering by U. Von Luxburg
B-Matching for Spectral Clustering by Jebara and Shchogolev
Geometry, Flows, and Graph-Partitioning Algorithms by Arora, Rao and Vazirani

Topic 12: Boosting
A Short Introduction to Boosting by Y. Freund and R. Schapire
Rapid Object Detection using a Boosted Cascade of Simple Features by P. Viola and M. Jones