A neural network modeled on the human brain successfully completed cognitive tasks
The authors of the project, scientists from the Montreal Hospital Neurological Institute and the Quebec Institute for Artificial Intelligence, conducted a thorough study of magnetic resonance imaging data in a large Open Science repository, reconstructed the patterns of brain connections and designed a neural network based on them. The result was a system of input-output modules resembling a biological brain. The functionality of the system was tested with cognitive tasks related to memory. This is not the usual beat-by-beat data fixation, but the interpretation of received information and its assimilation as the acquisition of skills - modeled on the work of the human brain. "The project combines two powerful scientific disciplines. Neuroscience and artificial intelligence share common roots but have recently diverged. Using artificial neural networks will help us understand how the structure of the brain enables its functions. In turn, using empirical data to create neural networks will open up design principles to create better artificial intelligence. Thus, these fields will help each other and enrich our understanding of the brain," commented one of the authors, Bratislav Misic, on his research. Translated with www.DeepL.com/Translator (free version)
2021-08-25