Project description
Brain networks useful for artificial intelligence
Artificial neural networks (ANNs) hold great opportunities and possibilities for advancements for many industries and services, nations, and organizations, from improved automation to ease of use and more efficiency. However, ANNs require extensive training and large amounts of data to accomplish even the most simplest of functions. By contrast, biological neural networks have evolved in the brain to perform highly complex functions efficiently. The ERC-funded ConnectomesToANNs project will provide several computational tools for capturing basic biological principles from reconstructions of neural networks in the brain in the design of ANNs. These 'biological' ANNs will improve the performance of artificial intelligence applications, while reducing training and data requirements.
Objective
Artificial neural networks (ANNs) have found applications in a wide variety of real-world problems. Despite this tremendous success, artificial intelligence systems still face major challenges due to their reliance on extensive training and large datasets. Recent reports indicate that the architecture of ANNs could be a prime target for reducing their training and data requirements.
We hypothesize that such architectural features can be identified from neuronal networks in the brain, which have evolved to efficiently perform highly specialized functions. Recent advances in electron microscopy will soon provide detailed reconstructions of large-scale neuronal networks from different brain areas, species, developmental stages and/or pathological conditions. However, even if such data become available, directly transforming neuronal network reconstructions into ANNs will raise problems of interpretability, due to their enormous complexity, and generalizability, due to high inter-individual variability.
Here, we will resolve these challenges by implementing a set of computational approaches that allow the extraction of rules that explain the wiring properties underlying dense connectomics data, the transfer of these anatomical principles into the design of ANN architectures, and the evaluation of how these principles impact performance on a battery of deep learning tasks. This unique methodology will lay the foundation for groundbreaking insights into how different network architectures facilitate specific brain functions, and also how the underlying anatomical principles can inform the development of more effective and efficient artificial intelligence systems.
Our methodology will be publicly accessible online to scientists, but also to companies and non-profit organizations that seek to improve the performance or reduce training data requirements for applications of deep learning.
Fields of science
- natural sciencesphysical sciencesopticsmicroscopyelectron microscopy
- natural sciencescomputer and information sciencesartificial intelligencefrugal artificial intelligence
- natural sciencescomputer and information sciencesartificial intelligencemachine learningdeep learning
- natural sciencescomputer and information sciencesartificial intelligencecomputational intelligence
Keywords
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Funding Scheme
HORIZON-ERC-POC - HORIZON ERC Proof of Concept GrantsHost institution
80539 Munchen
Germany