Active Bipartite Ranking
By James Cheshire

James Cheshire will give a talk on active bipartite ranking

Abstract

In this presentation, we will consider an active learning framework for the bipartite ranking problem. Motivated by numerous applications, ranging from supervised anomaly detection to credit-scoring through the design of medical diagnosis support systems, and usually formulated as the problem of optimizing (a scalar summary of) the ROC curve, bipartite ranking has been the subject of much attention in the passive context. Various dedicated algorithms have been recently proposed and studied by the machine-learning community. In contrast, active bipartite ranking rule is poorly documented in the literature. Due to its global nature, this learning task is much more complex than binary classification, for which many active algorithms have been designed. In this presentation, we will describe a dedicated algorithm, active-rank which aims to minimise the distance between the ROC curve of the ranking function built and the optimal one, w.r.t. the $\sup$ norm. We will show that, for a fixed confidence level $\epsilon$ and probability $\delta$, active-rank is PAC$(\epsilon,\delta)$. In addition, we will provide a problem dependent upper bound on the expected sampling time of active-rank.

Biography

Since 2022 he is a postdoc at Telecom Paris in collaboration with Stephan Clémençon . He obtained his PhD from Otto von Guericke University Magdeburg under the supervision of Alexandra Carpentier. His main research interests are in multi armed bandits and reinforcement learning.