Einladung zum Vortrag im Kolloquium Technische Kybernetik

 Clinical-Data-Based
Optimal Structured Treatment Interruption Strategies for HIV:
A Reinforcement Learning Approach

Dr. Guy-Bart Stan
Department of Engineering · Control Group
University of Cambridge · Cambridge · UK

    Zeit: Dienstag, 17.11. 2009 · 16:00 Uhr
Ort: IST-Seminarraum 3.243 · Pfaffenwaldring 9 · Campus Stuttgart-Vaihingen

Abstract

This research addresses the problem of computing optimal Structured Treatment Interruption strategies (STI) for HIV infected patients. STI represent a class of treatments in which patients are cycled on and off drug therapy at specific time instants.
The problem that we consider consists in designing efficient drug-scheduling strategies, i.e., strategies which bring the immune system into a state that allows it to independently maintain immune control over the virus (i.e., without help from anti-HIV drugs).
The design of optimal STI strategies must be performed in such a way that this transfer to a drug-independent viral control situation is realised with as low as possible drug-related systemic effects for the patients.
In this presentation, we propose an optimal control formulation of the problem and show how a sub-optimal solution can be obtained using ideas stemming from dynamic programming. The proposed approach is capable of computing (close-to) optimal STI strategies directly from clinical data, without the need for identifying a mathematical model of the HIV infection dynamics a priori.
To support our claims, we report simulation results obtained by running a recently proposed batch-mode reinforcement learning algorithm, known as fitted Q iteration, on numerically generated data containing noise and outliers.

Biographical Information

Guy-Bart Stan received his Electrical Engineering degree (with a speciality in Electronics) from the University of Liège, Belgium, in 2000, and his Ph.D. degree in Applied Sciences from the same university in 2005.
After working at Philips Applied Technologies as Senior Digital Signal Processing Engineer in 2005, he moved to the University of Cambridge as a Marie-Curie Fellow in 2006. From 2006 to 2009, Guy worked as Research Associate in the Control Group of the Department of Engineering of the University of Cambridge. Since November 2009, he is a Lecturer in the Department of Bioengineering of Imperial College London.
Guy's main research interests are currently in the fields of Nonlinear Systems Analysis, Control and Systems Biology with a particular interest in disease control and the analysis and design of Complex Networks.