Learning without gradient descent encoded by the dynamics of a neurobiological model
This work addresses the fundamental limitation of needing large training datasets and gradient descent for machine learning, offering an alternative for researchers seeking more biologically plausible and unsupervised learning paradigms.
This paper introduces a novel machine learning approach that leverages a neurobiologically derived model of dynamic signaling and geometric network structure. It demonstrates that MNIST images can be uniquely encoded and classified with nearly state-of-the-art accuracy (unspecified exact number) without any training or gradient descent.
The success of state-of-the-art machine learning is essentially all based on different variations of gradient descent algorithms that minimize some version of a cost or loss function. A fundamental limitation, however, is the need to train these systems in either supervised or unsupervised ways by exposing them to typically large numbers of training examples. Here, we introduce a fundamentally novel conceptual approach to machine learning that takes advantage of a neurobiologically derived model of dynamic signaling, constrained by the geometric structure of a network. We show that MNIST images can be uniquely encoded and classified by the dynamics of geometric networks with nearly state-of-the-art accuracy in an unsupervised way, and without the need for any training.