Thermodynamics of sensing, correlations and computation
File(s)
Author(s)
Brittain, Rory A.
Type
Thesis
Abstract
In this thesis we discuss several aspects of the thermodynamics of information. In some cases this is motivated by its application to biological organisms. First we examine the interpretation of a quantity proposed to measure sensors and find that it is not a general measure of sensor quality in the steady state. Then we propose a biochemical version of the Szilard engine and use it as the basis for a simple feeding model of a micro-organism. We find that more complex environments can support more complex inference processes. Finally, we use a similar set-up to the feeding model to model a reversible chemical Turing machine. We show that an additional memory tape is required but for a different reason than previously suggested. Also, in order to discount edge costs, many machines must be run simultaneously.
Version
Open Access
Date Issued
2019-09
Date Awarded
2020-01
Copyright Statement
Creative Commons Attribution NonCommercial NoDerivatives Licence
Advisor
Jones, Nicholas
Ouldridge, Thomas
Publisher Department
Mathematics
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)