In this research paper, Numenta proposes a novel theoretical framework for understanding what the neocortex does and how it does it. The framework is based on grid cells and has significant implications for neuroscience and machine intelligence.
This paper demonstrates how the application of Numenta’s brain-inspired, sparse algorithms achieves more than 50x speed-up with no loss of accuracy.
This companion piece explains the Thousand Brains Theory of Intelligence, one of the big ideas introduced in the October 2018 research paper A Framework for Intelligence and Cortical Function Based on Grid Cells in the Neocortex by Jeff Hawkins, Marcus Lewis, Scott Purdy, Mirko Klukas, and Subutai Ahmad. Written by non-neuroscientists, it can be read as a standalone piece or as a primer for the full research paper.
Recent proposals suggest that the brain might use similar mechanisms to understand the structure of objects in diverse sensory modalities, including vision. In machine vision, object recognition given a sequence of sensory samples of an image is a challenging problem when the sequence does not follow a consistent, fixed pattern – yet this is something humans do naturally and effortlessly. We explore how grid cell-based path integration in a cortical network can support reliable recognition of objects given an arbitrary sequence of inputs.
The paper walks through our implementation of brain-like SDRs in practical systems as a proof of concept. We implemented a sparse layer that can be dropped into existing deep learning and convolutional networks. We then trained sparse networks with back propagation, validated them with popular datasets and tested their accuracy with noisy images and sounds.