Seminar: Advanced Topics in Machine Learning

Fall Semester 2014

Course Catalogue Info

Abstract. In this seminar, recent papers of the pattern recognition and machine learning literature are presented and discussed. Possible topics cover statistical models in computer vision, graphical models and machine learning. The seminar "Advanced Topics in Pattern Recognition" familiarizes students with recent developments in pattern recognition and machine learning. Original articles have to be presented and critically reviewed. The students will learn how to structure a scientific presentation in English which covers the key ideas of a scientific paper. An important goal of the seminar presentation is to summarize the essential ideas of the paper in sufficient depth while omitting details which are not essential for the understanding of the work. The presentation style will play an important role and should reach the level of professional scientific presentations. The seminar will cover a number of recent papers which have emerged as important contributions to the pattern recognition and machine learning literature. The topics will vary from year to year but they are centered on methodological issues in machine learning like new learning algorithms, ensemble methods or new statistical models for machine learning applications. Frequently, papers are selected from computer vision or bioinformatics - two fields, which relies more and more on machine learning methodology and statistical models.

Introduction Slides with general information and all topics

 

Seminar Hours:
Tue 16-18 CAB H 52
Thu 16-18 CHN G 22
Start of sessions: 14. Oct

 

Contact:

Professors: Joachim M. Buhmann, Thomas Hofmann, Andreas Krause
Postdocs: Hamed Hassani, Martin Jaggi, Brian McWilliams

 

Schedule:

Tuesday Schedule Thursday Schedule
date topic presenter paper date topic presenter paper
14-Oct A Eirini Representation Learning: A Review and New Perspectives 16-Oct H Matthias A Scalable Bootstrap for Massive Data
  A Yannic On Optimization Methods for Deep Learning H Nico Iterative discovery of multiple alternative clustering views
21-Oct A Bogdan Improving neural networks by preventing co-adaptation of feature detectors (+maybe Dropout Training as Adaptive Regularization) 23-Oct E Philipe Connecting the Dots Between News Articles
  A Marco Distributed Representations of Sentences and Documents E Ueli Beyond Keyword Search: Discovering Relevant Scientific Literature
28-Oct A Shuoran Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank 30-Oct E Timo Learning Mixtures of Submodular Shells with Application to Document Summarization
  A Nicolas Natural Language Processing (Almost) from Scratch F Rocco A Simple and Practical Algorithm for Differentially Private Data Release
4-Nov A Jeremias Learning Feature Representations with K-Means 6-Nov F Martin Iterative learning for reliable crowdsourcing systems
  B Lei Pegasos: Primal Estimated Sub-Gradient Solver for SVM (+maybe A simpler approach to obtaining an O(1/t) convergence rate for SGD) F Maria The Multidimensional Wisdom of Crowds
11-Nov B Aleksandar Stochastic Dual Coordinate Ascent Methods for Regularized Loss Minimization 13-Nov F Severin Modeling Information Propagation with Survival Theory
  F Marco Regret Minimization and the Price of Total Anarchy
18-Nov B Sandro Semi-Stochastic Gradient Descent Methods / Accelerating Stochastic Gradient Descent using Predictive Variance Reduction 20-Nov G Shiling A spectral algorithm for learning mixtures of distributions
  B Jonas Stochastic Optimization with Importance Sampling G Johannes A spectral algorithm for learning hidden Markov models
25-Nov D Frederick Factoring nonnegative matrices with linear programs 27-Nov I Leonhard A Unified Framework for Probabilistic Component Analysis
  D Ivo Robust Near-Separable Non-negative Matrix Factorization Using Linear Optimization I Yuhua Decision Jungles: Compact and Rich Models for Classification

 

Tuesday Topics:

A - Learning Representations / Deep Networks

B - Optimization

C - Hashing & Randomization

D - Matrix Factorization

Thursday Topics:

E - Data Summarization

F - Networks, Crowds & Privacy

G - Spectral Learning (quite theoretical)

H - Dependence and Model Validation

I - Structure Learning / Graphical Models