Speaker |
Trevor Cohn |
|---|---|
Affiliation |
University of Sheffield |
Date |
Friday, 28 February 2014 |
Time |
13:00-14:00 |
Location |
Zoom |
Link |
Malet Place Engineering 1.02 |
Event series |
DeepMind/ELLIS CSML Seminar Series |
Abstract |
Gaussian Processes are non-parametric Bayesian models which support flexible kernels and Bayesian reasoning under uncertainty. Despite their strengths and growing adoption in machine learning, they have had very few applications to language processing. In this talk I will outline my recent work which represent some the first applications of GPs to language, and show significant improvements beyond state of the art in several language processing tasks. The first task I consider is machine translation evaluation, a task made difficult due to individual annotators bringing different biases, interpretations of the task and levels of consistency. I show how this problem can be framed as regression using a multi-task GP prior, such that individual models are learned while explicitly learning correlations between these models. I will also present applications to social media, including user impact prediction and identification of temporal periodicities in text usage. In both cases Gaussian Processes allow for much more accurate and flexible modelling than alternative methods, raising questions about the near-universal use of linear models in NLP. |
Biography |