Implementing engrams from a machine learning perspective: the relevance of a latent space
This work provides an incremental theoretical analysis of brain-inspired models, relevant for neuroscience and AI researchers interested in cognitive architectures.
The paper examines the role of latent space dimensionality in autoencoders as models for brain engrams, linking it to species-specific connectome differences and cognitive capacities, while noting that machine learning systems are not constrained by such biological limitations.
In our previous work, we proposed that engrams in the brain could be biologically implemented as autoencoders over recurrent neural networks. These autoencoders would comprise basic excitatory/inhibitory motifs, with credit assignment deriving from a simple homeostatic criterion. This brief note examines the relevance of the latent space in these autoencoders. We consider the relationship between the dimensionality of these autoencoders and the complexity of the information being encoded. We discuss how observed differences between species in their connectome could be linked to their cognitive capacities. Finally, we link this analysis with a basic but often overlooked fact: human cognition is likely limited by our own brain structure. However, this limitation does not apply to machine learning systems, and we should be aware of the need to learn how to exploit this augmented vision of the nature.