Hierarchical Temporal Memory with Numenta Platform

The first release of the Numenta Platform for Intelligent Computing (NuPIC) is a research release targeted at sophisticated developers for the purpose of education and experimentation. NuPIC implements a hierarchical temporal memory system (HTM) patterned after the human neocortex. We expect NuPIC to be used on problems that, generally speaking, involve identifying patterns in complex data. The ultimate applications likely will include vision systems, robotics, data mining and analysis, and failure analysis and prediction.

Numenta is committed to creating and supporting an open, collaborative community of companies and individuals interested in working on HTM systems. Concurrent with the Numenta Platform release, Numenta also has launched developer community tools and training materials.

 

Numenta Technology

Numenta has built a new type of memory system, called Hierarchical Temporal Memory (HTM), modeled on Jeff Hawkins’ theory of how the human neocortex works. HTM is hierarchical because it consists of memory modules connected in a hierarchical fashion. The hierarchy resembles an inverted tree with many memory modules at the bottom of the hierarchy and fewer at the top. HTM is “temporal” because each memory module stores and recalls sequences of patterns. HTM is hierarchical both temporally and spatially.

An HTM system is not programmed in a traditional sense; instead it is trained. Sensory data is applied to the bottom of the hierarchy and the HTM system automatically discovers the underlying patterns in the sensory input. You might say it “learns” what objects are in the world and how to recognize them. Time is an essential element of how HTM systems work. First, to learn the patterns in the world, the sensory data must flow over time just as we move our eyes to see and move our hands to feel. Second, because every memory module stores sequences of patterns, HTM systems can be used to make predictions of the future. They not only discover and recognize objects but they can make predictions about how objects will behave going forward in time.