T-7: An Introduction to Bayesian Machine Learning for Multimedia Information Processing
Monday All-Day, June 23, 09:30 - 17:30
Presented by
A. Taylan Cemgil, University of Cambridge, UK
Abstract
Recently, there has been a significant growth in the number of multimedia information processing applications that employ ideas from statistical machine learning and probabilistic modeling. In this paradigm, multimedia data (music, audio, video, images, text) are viewed as realizations from highly structured stochastic processes. Once a model is constructed, several interesting problems such as transcription, coding, classification, restoration, tracking, source separation or resynthesis etc. can be formulated as Bayesian inference problems. In this context, graphical models provide a "language" to construct models for quantification of prior knowledge. Unknown parameters in this specification are estimated by probabilistic inference. Often, however, the problem size poses an important challenge and in order to render the approach feasible, specialized inference methods need to be tailored to improve the computational speed and efficiency. The scope of the proposed tutorial is as follows: First, we will review the fundamentals of probabilistic models, with some focus on music, video and text data. Then, we will discuss the numerical techniques for inference in these models. In particular, we will review both exact and approximate stochastic (Monte Carlo) and deterministic (variational) inference techniques. Our ultimate aim is to provide a basic understanding of probabilistic modeling for multimedia processing, associated computational techniques and a roadmap such that researchers in multimedia information processing new to the Bayesian approach can orient themselves in the relevant literature and understand the current state of the art.
Speaker Biography
A. Taylan Cemgil received his B.Sc. and M.Sc. in Computer Engineering, Bogazici University, Turkey and his Ph.D. (2004) from Radboud University Nijmegen, the Netherlands with a thesis on Bayesian music transcription. He worked as a postdoctoral researcher at the University of Amsterdam on vision based multi object tracking. He is currently a research associate at the Signal Processing and Communications Lab., University of Cambridge, UK, where he cultivates his interests in machine learning methods, stochastic processes and statistical signal processing. His research is focused towards devoloping computational techniques for audio, music and multimedia processing.