Automated Model Inference from Neural Dynamics for a Mechanistic Understanding of Cognition

Project Information

AutoMIND

Grant agreement ID: 101030918

DOI

10.3030/101030918

Project closed

EC signature date 21 April 2021

Start date 1 May 2021

End date 30 April 2023

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 162 806,40

EU contribution

€ 162 806,40

Coordinated by

EBERHARD KARLS UNIVERSITAET TUEBINGEN
Germany

Project description

Machine learning-based model for in-depth study of cognition

The introduction of mechanistic models well aligned with experimental data could significantly improve understanding of cellular and network properties underlying cognition, and what leads to pathological conditions. Funded by the Marie Skłodowska-Curie Actions programme, the AutoMIND project proposes to develop a machine learning-based model inference tool capable of identifying parameters of particular spiking neural network models while capturing arbitrary target neural dynamics from human brain recordings. The tool capitalises on recent advances in simulation-based inference techniques, incorporating simultaneous parameter manifold-learning and gradient-based simulations. Moreover it is capable of solving neuroscientific problems via in silico experiments.

Objective

Human cognition depends on complex coordinated dynamics of neural populations, which is shaped by a rich heterogeneity in cellular properties and network connectivity patterns of neural circuits. While cognitive neuroscience leverages macroscopic brain signals to relate neural activity to behavioral states, we currently cannot dissect them for their physiological contributions, hindering mechanistic interpretations of experimental data. One way to systematically study how physiological parameters shape neural dynamics is through mechanistic modeling of spiking neural networks. However, current modeling approaches are not quantitatively constrained by observed electrophysiological data, and often require painstaking and ad-hoc parameter-tuning by hand. Efficient discovery of mechanistic models that are consistent with experimental data would dramatically accelerate our understanding of how cellular and network properties impact cognition, and why it breaks down in pathological states, representing a radical departure from how neural data is analyzed in cognitive neuroscience. To this end, I propose to develop a machine learning-assisted model inference tool—Automated Model Inference from Neural Dynamics (AutoMIND)—that can identify parameters of candidate spiking neural network models that could capture arbitrary target neural dynamics from human brain recordings. AutoMIND extends on recent advances in simulation-based inference techniques, incorporating simultaneous parameter manifold-learning and gradient-based simulations. AutoMIND has broad utility for tackling neuroscientific questions by enabling expedited in-silico experiments. Here, I apply it to multiscale neural data to study how cellular and network properties shape: 1) the emergence of synchronous network oscillations during early neurodevelopment, and 2) the difference in neural dynamics and computation between sensory and association cortices—two questions of fundamental importance to neuroscience.

Fields of science (EuroSciVoc)

• natural sciences › biological sciences › neurobiology › cognitive neuroscience
• natural sciences › computer and information sciences › artificial intelligence › computational intelligence

Keywords

• simulation-based inference
• likelihood-free inference
• mechanistic modeling
• spiking neural networks

Programme(s)

• H2020-EU.1.3. - EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions MAIN PROGRAMME
  
• H2020-EU.1.3.2. - Nurturing excellence by means of cross-border and cross-sector mobility
  

Topic(s)

• MSCA-IF-2020 - Individual Fellowships
  

Funding Scheme

MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)

Call for proposal

H2020-MSCA-IF-2020

Coordinator

EBERHARD KARLS UNIVERSITAET TUEBINGEN

• Net EU contribution: € 162 806,40
• Address: GESCHWISTER-SCHOLL-PLATZ ; 72074 Tuebingen ; Germany
• Region: Baden-Württemberg › Tübingen › Tübingen, Landkreis
• Activity type: Higher or Secondary Education Establishments
• Total cost: € 162 806,40