Poster M6 - Artificial Brains for Artificial Intelligence: Dendritic Integration Inspired Neural Networks

Introduction: Artificial neural networks (ANNs) are capable of complex feature extraction and classification with applications in robotics, natural language processing, and data science. Yet, many ANNs have several key limitations; notably, current neural network architectures require enormous training datasets and are computationally inefficient. It has been posited that biophysical computations in single neurons of the brain can inspire computationally more efficient algorithms and neural network architectures. Recently, research on dendrites, including work from our lab, suggests that each biological neuron has a hidden subcellular computational power that is more complex than the conventional perceptron, providing a promising substrate for more efficient ANNs.

Materials and

Methods: we propose the Interconnected Dendritic Network (IDN), a new type of ANN that takes close inspiration from cellular and subcellular networks of pyramidal neurons. Each neuron in an IDN has a set of dendrites that receive inputs; these dendrites are subdivided into branches of multiple orders to mimic dendritic organization in biological neurons closely. We further employ a family of physiologically-inspired activation functions to characterize dendritic input-output transformations. Instead of discrete layers, neurons are arranged in an n-dimensional space following topographical connectivity rules, forming a recurrent network. The network comprises excitatory and inhibitory neurons acting in unison to regulate network activity. Learning happens by altering synaptic weights based on Hebbian plasticity, approximating synaptic weight distributions observed in biological networks.

Results, Conclusions, and Discussions: The IDN reached over 95% accuracy on digit classification after training on only 400 data points and utilizing 3.6% of computational operations compared to a traditional dense model performing the same task. Further, we show that such computational efficiency critically depends on the structure of the dendritic network. Thus, we present a model that performs in low-data applications, is computationally efficient, and utilizes biophysically grounded mechanisms to mitigate the limitations of current ANNs.

Acknowledgements (Optional):

Funding: This work was supported by the following grants to K.J. the National Institutes of Health, New Innovator Award NIH DP2MH136494. This material is also based upon work supported by the Air Force Office of Scientific Research under award numbers FA9550-22-1-0078 and FA9550-23-1-0701. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the United States Air Force. H.F.K. is supported by the NIH T32 – 5T32DC016853 and NSF GRFP (DGE-1842166).