Research
Machine learning methods for neuroscience
Advances in experimental methods are generating data at an unprecedented scale and resolution, at the single neuron, neural network, and behaviour levels. Extracting knowledge from this wealth of data increasingly depends on accurate, scalable and interpretable theoretical models of neural activity and behaviour. In our lab, we believe that such effort can be potentiated with modern machine learning methods.
Despite the tremendous advances in machine learning, most machine learning methods are not geared at generating interpretable insights. Our lab answers these needs by developing and applying machine learning methods for mechanistic insight in neuroscience. In close collaboration with experimental partners, we aim to design more accurate models, quantitatively test mechanistic hypotheses and derive experimentally-testable predictions, with the ultimate goal of refining our understanding of neural systems in health and disease.
Selected publications
Multifidelity simulation-based inference for computationally expensive simulators
Krouglova A.N., Johnson H.R., Confavreux B., Deistler M., Gonçalves P.J.
We introduce a new multifidelity simulation-based inference method that combines many inexpensive low-fidelity simulations with a small number of costly high-fidelity ones to efficiently estimate model parameters. Across benchmark and neuroscience tasks, this approach achieves comparable performance to current methods while requiring up to 100 times fewer high-fidelity simulations.
JAXLEY: differentiable simulation enables large-scale training of detailed biophysical models of neural dynamics
Deistler M., Kadhim K.L., Pals M., Beck J., Huang Z., Gloeckler M., Lappalainen J.K., Schröder C., Berens P., Gonçalves P.J., Macke J.H.
We have developed a powerful new software toolbox that brings the tools of modern machine learning to realistic models of brain activity. This open-source framework, called JAXLEY, combines the precision of biophysical models with the scalability and flexibility of modern machine learning techniques.
Energy efficient network activity from disparate circuit parameters
Deistler M., Macke J.H.*, Gonçalves P.J.*
Neural systems have the remarkable feature of showing similar activity patterns despite having disparate underlying mechanistic properties. This feature, called parameter degeneracy, underlies the capacity of neural systems to compensate for perturbations to their components. Less understood is whether parameter degeneracy is reduced (or eliminated) by biological constraints, notably to preserve metabolic efficiency or robustness to environmental fluctuations. Developing machine learning methods for degeneracy analysis, we investigated this question in a computational model of the pyloric circuit in the crab stomatogastric ganglion.
Training deep neural density estimators to identify mechanistic models of neural dynamics
Gonçalves P.J.*, Lueckmann J.*, Deistler M.*, Nonnenmacher M., Oecal K., Bassetto G., Chintaluri C., Podlaski W.F., Haddad S.A., Vogels T.P., Greenberg D.S., Macke J.H.
We designed an algorithm that makes it easier to fit mathematical models to experimental data. First, the algorithm trains an artificial neural network to predict which models are compatible with simulated data. After initial training, the method can rapidly be applied to either raw experimental data or selected data features. The algorithm then returns the models that generate the best match. This newly developed machine learning tool was able to automatically identify models which can replicate the observed data from a diverse set of neuroscience problems, and may help bridge the gap between ‘data-driven’ and ‘theory-driven’ approaches.
Sequential Neural Posterior Estimation or SNPE
In SNPE, parameters for a mechanistic model of neural activity are drawn from a prior distribution provided by the modeler. Running simulations using these parameters creates simulated data on which a deep neural network can be trained. After training, the network can transform experimental data into a posterior distribution indicating which parameter values are consistent with that data.
Can we do accurate inference with few simulations?
By combining cheap approximate simulations with a small number of expensive, accurate ones, we can achieve similar results as state-of-the-art simulation-based inference methods using up to 100x fewer costly simulations, making Bayesian inference much more practical for complex models.
Is the pyloric network of the crustacean stomatogastric ganglion robust to perturbations of its parameters?
It turns out that we can move between two data-compatible parameter configurations along a parameter path without ever leaving high probability terrain. This parameter path corresponds therefore to a direction of robustness to perturbations.