Deep reinforcement learning for optimal experimental design in biology.
The field of optimal experimental design uses mathematical techniques to determine experiments that are maximally informative from a given experimental setup. Here we apply a technique from artificial intelligence-reinforcement learning-to the optimal experimental design task of maximizing confidenc...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2022-11-01
|
| Series: | PLoS Computational Biology |
| Online Access: | https://doi.org/10.1371/journal.pcbi.1010695 |
| _version_ | 1797973976719294464 |
|---|---|
| author | Neythen J Treloar Nathan Braniff Brian Ingalls Chris P Barnes |
| author_facet | Neythen J Treloar Nathan Braniff Brian Ingalls Chris P Barnes |
| author_sort | Neythen J Treloar |
| collection | DOAJ |
| description | The field of optimal experimental design uses mathematical techniques to determine experiments that are maximally informative from a given experimental setup. Here we apply a technique from artificial intelligence-reinforcement learning-to the optimal experimental design task of maximizing confidence in estimates of model parameter values. We show that a reinforcement learning approach performs favourably in comparison with a one-step ahead optimisation algorithm and a model predictive controller for the inference of bacterial growth parameters in a simulated chemostat. Further, we demonstrate the ability of reinforcement learning to train over a distribution of parameters, indicating that this approach is robust to parametric uncertainty. |
| first_indexed | 2024-04-11T04:11:38Z |
| format | Article |
| id | doaj.art-368750c01d7144d797d7622f55dae466 |
| institution | Directory Open Access Journal |
| issn | 1553-734X 1553-7358 |
| language | English |
| last_indexed | 2024-04-11T04:11:38Z |
| publishDate | 2022-11-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS Computational Biology |
| spelling | doaj.art-368750c01d7144d797d7622f55dae4662023-01-01T05:31:13ZengPublic Library of Science (PLoS)PLoS Computational Biology1553-734X1553-73582022-11-011811e101069510.1371/journal.pcbi.1010695Deep reinforcement learning for optimal experimental design in biology.Neythen J TreloarNathan BraniffBrian IngallsChris P BarnesThe field of optimal experimental design uses mathematical techniques to determine experiments that are maximally informative from a given experimental setup. Here we apply a technique from artificial intelligence-reinforcement learning-to the optimal experimental design task of maximizing confidence in estimates of model parameter values. We show that a reinforcement learning approach performs favourably in comparison with a one-step ahead optimisation algorithm and a model predictive controller for the inference of bacterial growth parameters in a simulated chemostat. Further, we demonstrate the ability of reinforcement learning to train over a distribution of parameters, indicating that this approach is robust to parametric uncertainty.https://doi.org/10.1371/journal.pcbi.1010695 |
| spellingShingle | Neythen J Treloar Nathan Braniff Brian Ingalls Chris P Barnes Deep reinforcement learning for optimal experimental design in biology. PLoS Computational Biology |
| title | Deep reinforcement learning for optimal experimental design in biology. |
| title_full | Deep reinforcement learning for optimal experimental design in biology. |
| title_fullStr | Deep reinforcement learning for optimal experimental design in biology. |
| title_full_unstemmed | Deep reinforcement learning for optimal experimental design in biology. |
| title_short | Deep reinforcement learning for optimal experimental design in biology. |
| title_sort | deep reinforcement learning for optimal experimental design in biology |
| url | https://doi.org/10.1371/journal.pcbi.1010695 |
| work_keys_str_mv | AT neythenjtreloar deepreinforcementlearningforoptimalexperimentaldesigninbiology AT nathanbraniff deepreinforcementlearningforoptimalexperimentaldesigninbiology AT brianingalls deepreinforcementlearningforoptimalexperimentaldesigninbiology AT chrispbarnes deepreinforcementlearningforoptimalexperimentaldesigninbiology |