<ul><li>Self-supervised learning has become increasingly important in machine intelligence.</li><li>Meta-representational predictive coding (MPC) offers a neurobiologically plausible framework for self-supervised learning.</li><li>MPC sidesteps the need for learning a generative model of sensory input by learning to predict representations of the input.</li><li>MPC uses active inference to drive the learning of representations by sampling informative portions of the sensorium.</li></ul>

Meta-Representational Predictive Coding: Biomimetic Self-Supervised Learning

Discover more