This paper demonstrates how the application of Numenta’s brain-inspired, sparse algorithms achieves more than 100x speed-up on inference tasks compared to dense networks with no loss of accuracy.
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.
This paper shows that a set of grid cell modules, each with only 2D responses, can generate unambiguous and high-capacity representations of variables in much higher-dimensional spaces.
This paper explains how location signals can be generated with a location layer that utilizes grid-cell-like neurons. It builds on our previous paper, A Theory of How Columns in the Neocortex Enable Learning the Structure of the World.
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.