University of Cambridge > > Machine Learning @ CUED > Clinical data based optimal STI strategies for HIV: a reinforcement learning approach

Clinical data based optimal STI strategies for HIV: a reinforcement learning approach

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Carl Edward Rasmussen.

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 (without help from any drug) maintain immune control over the virus. Also, this transfer to a drug-independent viral control situation should be done with as low as possible drug-related systemic effects for the patients.

In this presentation, we show that reinforcement learning may be useful to extract (close-to) optimal STI strategies directly from clinical data, without the need of identifying a mathematical model of HIV infection dynamics. 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.

The corresponding paper can be found at

This talk is part of the Machine Learning @ CUED series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2024, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity