Programme

Gaussian Process Round Table Programme

---

The proposed schedule contains talks of 45 minutes length with fifteen minutes of discussion.

	Thursday 9th June
9:00	Coffee & Welcome
9:15	Tony O'Hagan	Abstract
10:15	Chris Williams	Abstract
11:15	Coffee Break
11:45	Kai Yu	Abstract
12:45	Lunch
13:45	Zoubin Ghahramani	Abstract
14:00	Ed Snelson	Abstract
15:45	Tea
16:15	Carl Rasmussen	Abstract
17:15	Discussion I

20:30 Dinner in Monsal Head Hotel

	Friday 10th June
9:00	Neil Lawrence	Abstract
9:30	Manfred Opper	Abstract
10:00	Ole Winther	Abstract
11:00	Coffee Break
12:00	Anton Schwaighofer	Abstract
13:00	Lunch
14:00	Lehel Csato	Abstract
15:00	Tony Dodd	Abstract
16:00	Tea
16:30	Joaquin Quinonero Candela	Abstract
17:30	Discussion II

 

---

Some Suggested Subject Matter for Discussion

If you haven't already decided what you are going to say, Carl and Joaquin have sent a list of issues that they would like to hear about. These are certainly things we should also discuss. If you are not giving a full talk you might like to bring some of your thoughts on some of these issues.

Good applications

Applications for which Gaussian processes are particularly suited, and seem to perform better than other alternative modelling approaches.

Optimization via ML vs cross-validation

Though in regression it seems that optimizing the marginal likelihood leads to good generalization performance, the same cannot be said for classification where at times maximizing the marginal likelihood makes the test error become worse.

Empirical comparisons

the question is whether there exists enough (if any) well designed empirical  comparisons that allow making assessments on the performance of Gaussian Processes compared to competing methods. If not, one may want to motivate the design of good empirical comparisons.

GPs and large scale datasets

what are the most effective means of dealing with large datasets? A number of methods have been proposed, which all seem to suffer from diverse limitations (stability, ability to optimize reduced sets and hyperparameters, quality of the predictive distributions, etc)

Covariance functions:

Stein's book "Interpolation of Spatial Data" claims that "the lengthscales" are not important, only the shape of the covariance function is. The ubiquitous squared exponential covariance function suffers from limitations (ie. it is too smooth). How much effort is yet to be devoted to investigating new covariance functions?

The non-Gaussian Case

In classification, as well as in regression with general noise models, analytic inference is impossible, and use is made of approximations. The number and variety of these is high, and no clear consensus seems to exist on which are better than others.