Dulwich Hill

I fell in love at the Dulwich golf club.

Probabilistic Hill-climbing

Many learning tasks involve searching through a discrete space of performance elements, seeking an element whose future utility is expected to be high. As the task of nding the global optimum is often intractable, many practical learning systems use simple forms of hill-climbing to nd a locally optimal element. However, hill-climbing can be complicated by the fact that the utility value of a pe...

Sarah Amy Hill

When Jim Lyons-Weiler and I discussed what we might do with an Open Access journal on Cancer Informatics, we both felt that it was important to avoid loosing focus on what the end-point of both the research and its dissemination is. In all its forms cancer is inevitably an unpleasant affl iction that harms every aspect of the lives of sufferers and those who love them. Jim and I want this journ...

A Multivariate Hill Estimator

We propose a simple and semi-parametric estimator for the tail index of a regular varying elliptical random vector. Since, for univariate random variables, our estimator boils down to the Hill estimator and it inherits the simplicity and asymptotic properties, we name it after Bruce M. Hill. The estimator is based on the distance between an elliptical probability contour and the outer – or exce...

Stochastic Enforced Hill-Climbing

Enforced hill-climbing is an effective deterministic hillclimbing technique that deals with local optima using breadth-first search (a process called “basin flooding”). We propose and evaluate a stochastic generalization of enforced hill-climbing for online use in goal-oriented probabilistic planning problems. We assume a provided heuristic function estimating expected cost to the goal with fla...