A reaction network scheme for hidden Markov model parameter learning
Authors:
Open Access
- 1 June 2023
- journal article
- research article
- Published by The Royal Society in Journal of The Royal Society Interface
- Vol. 20 (203) , 20220877
Abstract
With a view towards artificial cells, molecular communication systems, molecular multiagent systems and federated learning, we propose a novel reaction network scheme (termed the Baum–Welch (BW) reaction network) that learns parameters for hidden Markov models (HMMs). All variables including inputs and outputs are encoded by separate species. Each reaction in the 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 positive fixed point of the BW algorithm for HMMs is a fixed point of the reaction network scheme, and vice versa. Furthermore, we prove that the ‘expectation’ step and the ‘maximization’ step of the reaction network separately converge exponentially fast and compute the same values as the E-step and the M-step of the BW algorithm. We simulate example sequences, and show that our reaction network learns the same parameters for the HMM as the BW algorithm, and that the log-likelihood increases continuously along the trajectory of the reaction network.Keywords
All Related Versions
Funding Information
- INSPIRE DST Scholarship for Higher Education
- Novo Nordisk Fonden (NNF19OC0058354)
- INSPIRE DST Scholarship for Higher Education
- Novo Nordisk Fonden (NNF19OC0058354)
This publication has 40 references indexed in Scilit:
- Deterministic function computation with chemical reaction networksNatural Computing, 2013
- Evolution of Associative Learning in Chemical NetworksPLOS Computational Biology, 2012
- Identifying sources of variation and the flow of information in biochemical networksProceedings of the National Academy of Sciences, 2012
- Abstractions for DNA circuit designJournal of The Royal Society Interface, 2011
- Neural network computation with DNA strand displacement cascadesNature, 2011
- DNA as a universal substrate for chemical kineticsProceedings of the National Academy of Sciences, 2010
- Computing Algebraic Functions with Biochemical Reaction NetworksArtificial Life, 2009
- Training hidden Markov models with multiple observations-a combinatorial methodIEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
- Hidden Markov Models for Speech RecognitionTechnometrics, 1991
- A tutorial on hidden Markov models and selected applications in speech recognitionProceedings of the IEEE, 1989