Bayesian optimization of distributed neurodynamical controller models for spatial navigation
Dynamical systems models for controlling multi-agent swarms have demonstrated advances toward resilient, decentralized navigation algorithms. We previously introduced the NeuroSwarms controller, in which agent-based interactions were modeled by analogy to neuronal network interactions, including att...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-09-01
|
| Series: | Array |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590005622000601 |
| _version_ | 1797999029283454976 |
|---|---|
| author | Armin Hadzic Grace M. Hwang Kechen Zhang Kevin M. Schultz Joseph D. Monaco |
| author_facet | Armin Hadzic Grace M. Hwang Kechen Zhang Kevin M. Schultz Joseph D. Monaco |
| author_sort | Armin Hadzic |
| collection | DOAJ |
| description | Dynamical systems models for controlling multi-agent swarms have demonstrated advances toward resilient, decentralized navigation algorithms. We previously introduced the NeuroSwarms controller, in which agent-based interactions were modeled by analogy to neuronal network interactions, including attractor dynamics and phase synchrony, that have been theorized to operate within hippocampal place-cell circuits in navigating rodents. This complexity precludes linear analyses of stability, controllability, and performance typically used to study conventional swarm models. Further, tuning dynamical controllers by manual or grid-based search is often inadequate due to the complexity of objectives, dimensionality of model parameters, and computational costs of simulation-based sampling. Here, we present a framework for tuning dynamical controller models of autonomous multi-agent systems with Bayesian optimization. Our approach utilizes a task-dependent objective function to train Gaussian process surrogate models to achieve adaptive and efficient exploration of a dynamical controller model’s parameter space. We demonstrate this approach by studying an objective function selecting for NeuroSwarms behaviors that cooperatively localize and capture spatially distributed rewards under time pressure. We generalized task performance across environments by combining scores for simulations in multiple mazes with distinct geometries. To validate search performance, we compared high-dimensional clustering for high- vs. low-likelihood parameter points by visualizing sample trajectories in 2-dimensional embeddings. Our findings show that adaptive, sample-efficient evaluation of the self-organizing behavioral capacities of complex systems, including dynamical swarm controllers, can accelerate the translation of neuroscientific theory to applied domains. |
| first_indexed | 2024-04-11T10:58:06Z |
| format | Article |
| id | doaj.art-f741fc287caa4e4fb4a67cdec467e959 |
| institution | Directory Open Access Journal |
| issn | 2590-0056 |
| language | English |
| last_indexed | 2024-04-11T10:58:06Z |
| publishDate | 2022-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Array |
| spelling | doaj.art-f741fc287caa4e4fb4a67cdec467e9592022-12-22T04:28:43ZengElsevierArray2590-00562022-09-0115100218Bayesian optimization of distributed neurodynamical controller models for spatial navigationArmin Hadzic0Grace M. Hwang1Kechen Zhang2Kevin M. Schultz3Joseph D. Monaco4The Johns Hopkins University Applied Physics Laboratory, Laurel, 20723, MD, USA; Corresponding author.The Johns Hopkins University Applied Physics Laboratory, Laurel, 20723, MD, USA; Kavli Neuroscience Discovery Institute, Johns Hopkins University, Baltimore, 21218, VA, USADepartment of Biomedical Engineering, Johns Hopkins University School of Medicine, Baltimore, 21205, MD, USAThe Johns Hopkins University Applied Physics Laboratory, Laurel, 20723, MD, USADepartment of Biomedical Engineering, Johns Hopkins University School of Medicine, Baltimore, 21205, MD, USADynamical systems models for controlling multi-agent swarms have demonstrated advances toward resilient, decentralized navigation algorithms. We previously introduced the NeuroSwarms controller, in which agent-based interactions were modeled by analogy to neuronal network interactions, including attractor dynamics and phase synchrony, that have been theorized to operate within hippocampal place-cell circuits in navigating rodents. This complexity precludes linear analyses of stability, controllability, and performance typically used to study conventional swarm models. Further, tuning dynamical controllers by manual or grid-based search is often inadequate due to the complexity of objectives, dimensionality of model parameters, and computational costs of simulation-based sampling. Here, we present a framework for tuning dynamical controller models of autonomous multi-agent systems with Bayesian optimization. Our approach utilizes a task-dependent objective function to train Gaussian process surrogate models to achieve adaptive and efficient exploration of a dynamical controller model’s parameter space. We demonstrate this approach by studying an objective function selecting for NeuroSwarms behaviors that cooperatively localize and capture spatially distributed rewards under time pressure. We generalized task performance across environments by combining scores for simulations in multiple mazes with distinct geometries. To validate search performance, we compared high-dimensional clustering for high- vs. low-likelihood parameter points by visualizing sample trajectories in 2-dimensional embeddings. Our findings show that adaptive, sample-efficient evaluation of the self-organizing behavioral capacities of complex systems, including dynamical swarm controllers, can accelerate the translation of neuroscientific theory to applied domains.http://www.sciencedirect.com/science/article/pii/S2590005622000601Bayesian optimizationMulti-agent controlSwarmingDynamical systems modelsSpatial navigationUMAP |
| spellingShingle | Armin Hadzic Grace M. Hwang Kechen Zhang Kevin M. Schultz Joseph D. Monaco Bayesian optimization of distributed neurodynamical controller models for spatial navigation Array Bayesian optimization Multi-agent control Swarming Dynamical systems models Spatial navigation UMAP |
| title | Bayesian optimization of distributed neurodynamical controller models for spatial navigation |
| title_full | Bayesian optimization of distributed neurodynamical controller models for spatial navigation |
| title_fullStr | Bayesian optimization of distributed neurodynamical controller models for spatial navigation |
| title_full_unstemmed | Bayesian optimization of distributed neurodynamical controller models for spatial navigation |
| title_short | Bayesian optimization of distributed neurodynamical controller models for spatial navigation |
| title_sort | bayesian optimization of distributed neurodynamical controller models for spatial navigation |
| topic | Bayesian optimization Multi-agent control Swarming Dynamical systems models Spatial navigation UMAP |
| url | http://www.sciencedirect.com/science/article/pii/S2590005622000601 |
| work_keys_str_mv | AT arminhadzic bayesianoptimizationofdistributedneurodynamicalcontrollermodelsforspatialnavigation AT gracemhwang bayesianoptimizationofdistributedneurodynamicalcontrollermodelsforspatialnavigation AT kechenzhang bayesianoptimizationofdistributedneurodynamicalcontrollermodelsforspatialnavigation AT kevinmschultz bayesianoptimizationofdistributedneurodynamicalcontrollermodelsforspatialnavigation AT josephdmonaco bayesianoptimizationofdistributedneurodynamicalcontrollermodelsforspatialnavigation |