Biologically-Inspired Spatial Neural Networks
This work addresses interpretability and continual learning in neural networks for AI researchers, though it is incremental as it builds on existing biologically-inspired models.
The paper tackles the problem of designing neural networks that mimic biological systems by incorporating spatial positions and connection costs, resulting in neurons naturally clustering into task-specific groups when handling multiple tasks.
We introduce bio-inspired artificial neural networks consisting of neurons that are additionally characterized by spatial positions. To simulate properties of biological systems we add the costs penalizing long connections and the proximity of neurons in a two-dimensional space. Our experiments show that in the case where the network performs two different tasks, the neurons naturally split into clusters, where each cluster is responsible for processing a different task. This behavior not only corresponds to the biological systems, but also allows for further insight into interpretability or continual learning.