A Reaction Network Scheme Which Implements Inference and Learning for Hidden Markov Models

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

With a view towards molecular communication systems and molecular multi-agent systems, we propose the Chemical Baum-Welch Algorithm, a novel reaction network scheme that learns parameters for Hidden Markov Models (HMMs). Each reaction in our scheme changes only one molecule of one species to one molecule of another. The reverse change is also accessible but via a different set of enzymes, in a design reminiscent of futile cycles in biochemical pathways. We show that every fixed point of the Baum-Welch algorithm for HMMs is a fixed point of our reaction network scheme, and every positive fixed point of our scheme is a fixed point of the Baum-Welch algorithm. We prove that the “Expectation� step and the “Maximization� step of our reaction network separately converge exponentially fast. We simulate mass-action kinetics for our network on an example sequence, and show that it learns the same parameters for the HMM as the Baum-Welch algorithm.

Original languageEnglish
Title of host publicationDNA Computing and Molecular Programming - 25th International Conference, DNA 25, Proceedings
EditorsYan Liu, Chris Thachuk
Number of pages26
PublisherSpringer
Publication date2019
Pages54-79
ISBN (Print)9783030268060
ISBN (Electronic)9783030268077
DOIs
Publication statusPublished - 2019
Event25th International Conference on DNA Computing and Molecular Programming, DNA 2019 - Seattle, United States
Duration: 5 Aug 20199 Aug 2019

Conference

Conference25th International Conference on DNA Computing and Molecular Programming, DNA 2019
LandUnited States
BySeattle
Periode05/08/201909/08/2019
SponsorInternational Society of Nanoscale Science, Computing, and Engineering, Microsoft Research (USA), National Science Foundation (USA), University of Washington
SeriesLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11648 LNCS
ISSN0302-9743

Links

ID: 226950914