TALK    Probabilistic Latent Tensor Factorisation

Date released: February 26, 2013

  •  TALK    Probabilistic Latent Tensor Factorisation
  • Date & Time:

    Tuesday, February 26, 2013; 12:00 PM

  • Abstract:

    Algorithms for decompositions of matrices are of central importance in machine learning, signal processing and information retrieval, with SVD and NMF (Nonnegative Matrix Factorisation) being the most widely used examples. Probabilistic interpretations of matrix factorisation models are also well known and are useful in many applications (Salakhutdinov and Mnih 2008; Cemgil 2009; Fevotte et. al. 2009). In the recent years, decompositions of multiway arrays, known as tensor factorisations have gained significant popularity for the analysis of large data sets with more than two entities (Kolda and Bader, 2009; Cichocki et. al. 2008). We will discuss a subset of these models from a statistical modelling perspective, building upon probabilistic Bayesian generative models and generalised linear models (McCulloch and Nelder). In both views, the factorisation is implicit in a well-defined hierarchical statistical model and factorisations can be computed via maximum likelihood.

    We express a tensor factorisation model using a factor graph and the factor tensors are optimised iteratively. In each iteration, the update equation can be implemented by a message passing algorithm, reminiscent to variable elimination in a discrete graphical model. This setting provides a structured and efficient approach that enables very easy development of application specific custom models, as well as algorithms for the so called coupled (collective) factorisations where an arbitrary set of tensors are factorised simultaneously with shared factors. Extensions to full Bayesian inference for model selection, via variational approximations or MCMC are also feasible. Well known models of multiway analysis such as Nonnegative Matrix Factorisation (NMF), Parafac, Tucker, and audio processing (Convolutive NMF, NMF2D, SF-SSNTF) appear as special cases and new extensions can easily be developed. We will illustrate the approach with applications in link prediction and audio and music processing.

  • Speaker:

    Prof. Taylan Cemgil
    Bogazici University, Istanbul, Turkey

    Taylan Cemgil received his Ph.D. (2004) from SNN, Radboud University Nijmegen, the Netherlands. Between 2004 and 2008 he worked as a postdoctoral researcher at the Signal Processing and Communications Lab., University of Cambridge, UK. He is currently an associate professor of Computer Engineering at Bogazici University, Istanbul, Turkey. He is interested in Bayesian statistical methods, machine learning and signal processing, in particulat audio and music processsing.

  • MERL Host:

    Jonathan Le Roux

  • External Link:

  • Research Area:

    Speech & Audio