Schedule

The event starts at 9:30am and ends at 6pm on Wednesday, August 28th, 2013. Registration opens at 9:00am.

The program includes

3 invited talks by Michael Jordan, Trevor Hastie and Yoram Singer;
6 short talks and a poster session;
and, most importantly, break/lunch time for discussion.

9:00am

Registration opens

9:25am

Welcome to BayLearn 2013!

9:30am
Keynote 1 Michael Jordan, Berkeley
On the Computational and Statistical Interface and "Big Data" (slides)

10:20am
Session 1 Xinghao Pan, Berkeley; Joseph Gonzalez; Stefanie Jegelka; Tamara Broderick; Michael Jordan
Optimistic Concurrency Control for Distributed Unsupervised Learning

Ameet Talwalkar, Berkeley
MLI: An API for Distributed Machine Learning

10:50am
Break

11:20am
Keynote 2 Trevor Hastie, Stanford
Sparse Linear Models (slides)
12:10pm
Session 2 Jon Malmaud, MIT; Kevin Murphy, Google
Extracting Entities and Relations from Web Tables Using a Non-parametric Generative Model

Manfred Warmuth, UC Santa Cruz; S.V.N. Vishwanathan, Purdue University; Nan Ding, Google
From Relative Entropies to Bregman Divergences, to the Design of Convex and Tempered Non-convex Losses

12:40pm
Lunch

2:00pm
Keynote 3 Yoram Singer, Google BOOM: BOOsting with Momentum (50min)

2:50pm
Session 3 Andrew Saxe, Stanford University; James McClelland; Surya Ganguli
Exact solutions to the nonlinear dynamics of learning in deep linear neural networks

Sida Wang, Stanford University; Stefan Wager; Percy Liang
Semi-supervised Dropout Training

3:20pm
Break

3:50pm
Posters Session with posters from and presentations from Session 1, 2, 3 (130min)

6:00pm

End of the Symposium

Posters

Lightning-Fast, Dirt-Cheap Parallel Stochastic Gradient Descent for Big Data in GLADE

Chengjie Qin, Florin Rusu

Hierarchical Multiclass Active Learning with Label Tree Embedding
Jiang Wang, Yu Cheng, Ying Wu

A Simple Assignment Model with Laplacian Smoothing
Miguel Carreira-Perpinan, Weiran Wang

What You Hear Is What You Get: Audio-Based Video Content Analysis
Benjamin Elizalde, Karl Ni, Gerald Friedland

Discriminative Boltzmann Machines for Structured Output Prediction
Jimei Yang, Simon Safar, Ming-Hsuan Yang

Inverse Covariance Estimation from Data with Missing Values using the Concave-Convex procedure
Timothy Hunter, Jerome Thai, Kene Akametalu

Using boosted Alternating Decision Trees to model cell type specific gene regulation
Sofia Kyriazopoulou Panagiotopoulou, Anshul Kundaje, Serafim Batzoglou

Semi-Supervised Learning for Integration of Aerosol Predictions from Multiple Satellite Instruments
Nemanja Djuric, Lakesh Kansakar, Slobodan Vucetic

Beyond the Netflix Prize: Recommender Systems for Ranking
Harald Steck

Fast Algorithm for Utility Maximization in Integer Programming
Guocong Song

Linear-time Training of Nonlinear Low-Dimensional Embeddings
Max Vladymyrov, Miguel Carreira-Perpinan

An Efficient Parameter Filtering Algorithm
Lei Li, Yusuf Bugra Erol, Bharath Ramsundar, Stuart Russell

From subspace clustering to full-rank matrix completion
Emmanuel Candes, Lester Mackey, Mahdi Soltanolkotabi

Machine Learning in an Auction Environment
Patrick Hummel, R. Preston McAfee, Martin Zinkevich

Sparse multitask regression for predicting cancer drug sensitivity
David Knowles, Alexis Battle, Daphne Koller