Departmental Seminars http://www.cs.ox.ac.uk/seminars/

This is a series of non-specialist lectures which should be of interest to most members of the Department of Computer Science, as well as to other members of the University.

Everyone is very welcome to come, and to suggest new topics and speakers to the co-ordinator.

Tuesdays at 4.30pm in the Lecture Theatre B (unless otherwise stated)

en-gb 360 Sun, 19 May 2013 23:19:12 GMT Sun, 19 May 2013 23:19:12 GMT departmental http://blogs.law.harvard.edu/tech/rss/ Bayesian Learning via Stochastic Gradient Langevin Dynamics. Yee Whye Teh (Statistics Department, University of Oxford) http://www.cs.ox.ac.uk/seminars/864.html <p>The Bayesian approach to machine learning is a theoretically well-motivated framework to learning from data. It provides a coherent framework to reasoning about uncertainties, and an inbuilt protection against overfitting. However, computations in the framework can be expensive, and most approaches to Bayesian computations do not scale well to the big data setting. In this talk we propose a new computational approach for Bayesian learning from large scale datasets based on iterative learning from small mini-batches. By adding the right amount of noise to a standard stochastic gradient optimization algorithm we show that the iterates will converge to samples from the true posterior distribution as we anneal the stepsize. We apply the method to logistic regression and latent Dirichlet allocation, showing state-of-the-art performance.</p> <p>Joint work with Max Welling and Sam Patterson.</p> <br/>Lecture Theatre B Departmental Seminars oucl-seminar-departmental-864 Tue, 29 Jan 2013 12:21:30 GMT