Evolutionary algorithm optimization of biological learning parameters in a biomimetic neuroprosthesis

Abstract

Biomimetic simulation permits neuroscientists to better understand the complex neuronal dynamics of the brain. Embedding a biomimetic simulation in a closed-loop neuroprosthesis, which can read and write signals from the brain, will permit applications for amelioration of motor, psychiatric, and memory-related brain disorders. Biomimetic neuroprostheses require real-time adaptation to changes in the external environment, thus constituting an example of a dynamic data-driven application system. As model fidelity increases, so does the number of parameters and the complexity of finding appropriate parameter configurations. Instead of adapting synaptic weights via machine learning, we employed major biological learning methods: spike-timing dependent plasticity and reinforcement learning. We optimized the learning metaparameters using evolutionary algorithms, which were implemented in parallel and which used an island model approach to obtain sufficient speed. We employed these methods to train a cortical spiking model to utilize macaque brain activity, indicating a selected target, to drive a virtual musculoskeletal arm with realistic anatomical and biomechanical properties to reach to that target. The optimized system was able to reproduce macaque data from a comparable experimental motor task. These techniques can be used to efficiently tune the parameters of multiscale systems, linking realistic neuronal dynamics to behavior, and thus providing a useful tool for neuroscience and neuroprosthetics.

Multiscale dynamics and information flow in a data-driven model of the primary motor cortex microcircuit

“…5). All 15 strongest information flow projections targeted L5 cells, and exhibited lower mean peak frequencies when originating from upper layers (L2/3 and L4; 32-34 Hz) than when originating from deeper layers (L5 and L6; [35][36]. This analysis extends the microcircuit description by providing specifics about the cell classes, sublaminar regions and oscillation frequencies.…”

“…This in silico testbed can be systematically probed to study microcircuit dynamics, information flow and biophysical mechanisms with a level of resolution and precision not available experimentally. Unraveling the non-intuitive multiscale interactions occurring in M1 circuits will help us understand disease and develop new pharmacological and neurostimulation treatments for motor disorders 101,100,97,36,7,138,50,12,119 , and improve decoding methods for brain-machine interfaces 22,124,35,67 .…”

“…Power spectral density (PSD) of LFP and firing rates during spontaneous activity A. LFP PSDs at different depths for base model simulation: peaks were in high beta / low gamma range(28)(29)(30)(31)(32)(33)(34)(35). B. LFP PSDs across full model set (N=25) at upper L5B with mean (blue line) and interquartile ranges (bars).…”

NetPyNE: a tool for data-driven multiscale modeling of brain circuits

“…53,54 NetPyNE provides built-in support for several automated methods that have been successfully applied to both single cell and network optimization: grid-search and evolutionary algorithms. 2,[55][56][57][58][59][60] Grid search refers to evaluating all combinations of a fixed set of values for each parameter, resulting in a grid-like sampling of the multidimensional parameter space. Evolutionary algorithms (EAs) employ principles derived from evolution -selection, reproduction and mutation -to iteratively improve the model solutions.…”

Open Source Brain: a collaborative resource for visualizing, analyzing, simulating and developing standardized models of neurons and circuits

“…Standardized models currently on OSB OSB currently hosts standardized curated models from multiple regions of the brain including neocortex 8,9,21,[33][34][35][36][37][38] , cerebellum 10,[39][40][41] , hippocampus 7,[42][43][44] and olfactory bulb 45 . These include single cell models from the Allen Institute Cell Types Database 19 and the Blue Brain Project 8 , which have been converted to NeuroML, to enable visualization and simulation on OSB.…”