Exploring Parameter and Hyper-Parameter Spaces of Neuroscience Models on High Performance Computers With Learning to Learn

Abstract

Neuroscience models commonly have a high number of degrees of freedom and only specific regions within the parameter space are able to produce dynamics of interest. This makes the development of tools and strategies to efficiently find these regions of high importance to advance brain research. Exploring the high dimensional parameter space using numerical simulations has been a frequently used technique in the last years in many areas of computational neuroscience. Today, high performance computing (HPC) can provide a powerful infrastructure to speed up explorations and increase our general understanding of the behavior of the model in reasonable times. Learning to learn (L2L) is a well-known concept in machine learning (ML) and a specific method for acquiring constraints to improve learning performance. This concept can be decomposed into a two loop optimization process where the target of optimization can consist of any program such as an artificial neural network, a spiking network, a single cell model, or a whole brain simulation. In this work, we present L2L as an easy to use and flexible framework to perform parameter and hyper-parameter space exploration of neuroscience models on HPC infrastructure. Learning to learn is an implementation of the L2L concept written in Python. This open-source software allows several instances of an optimization target to be executed with different parameters in an embarrassingly parallel fashion on HPC. L2L provides a set of built-in optimizer algorithms, which make adaptive and efficient exploration of parameter spaces possible. Different from other optimization toolboxes, L2L provides maximum flexibility for the way the optimization target can be executed. In this paper, we show a variety of examples of neuroscience models being optimized within the L2L framework to execute different types of tasks. The tasks used to illustrate the concept go from reproducing empirical data to learning how to solve a problem in a dynamic environment. We particularly focus on simulations with models ranging from the single cell to the whole brain and using a variety of simulation engines like NEST, Arbor, TVB, OpenAIGym, and NetLogo.

Vast Parameter Space Exploration of the Virtual Brain: A Modular Framework for Accelerating the Multi-Scale Simulation of Human Brain Dynamics

“…In this regard, the utilization of specialized tools becomes pressing. One such tool that holds promise in navigating these immense data-sets is "Learning to Learn" (L2L) [37]. This automated machine learning framework is purpose-built for high-performance computing environments and is adept at employing gradient or evolutionary strategies to traverse expansive data generated by our framework.…”

HCEs‐Net: Hepatic cystic echinococcosis classification ensemble model based on tree‐structured Parzen estimator and snap‐shot approach

“…Traditionally, hyper‐parameter optimization has been done manually because it is too costly to search for hyper‐parameter automatically 25 . Simple grid search and random search are inadequate for hyper‐parameter search in deep learning models 26 .…”

“…Traditionally, hyper-parameter optimization has been done manually because it is too costly to search for hyper-parameter automatically. 25 Simple grid search and random search are inadequate for hyper-parameter search in deep learning models. 26 The modern Bayesian optimization can judge the following search direction according to the search parameters and improve search efficiency.…”

Efficient parameter calibration and real-time simulation of large-scale spiking neural networks with GeNN and NEST

“…Thus, both in scientific projects and for the development of real-world applications of SNNs, model tuning through parameter search typically creates the highest demand in computing time. Currently available methods (Feurer and Hutter, 2019) such as grid search or random search (LaValle et al, 2004;Bergstra and Bengio, 2012), Bayesian optimization (Parsa et al, 2019), and machine learning approaches (Carlson et al, 2014;Yegenoglu et al, 2022) require extensive sampling of the parameter space. We therefore suggest to include the parameter search in benchmarking approaches to the efficient simulation of large-scale SNNs.…”