Read Full Article
Welcome to NuPIC
Open Source Platform and Community for Machine Intelligence based on HTM Theory
NuPIC is an open source project based on a theory of neocortex called Hierarchical Temporal Memory (HTM). Parts of HTM theory have been implemented, tested, and used in applications, and other parts of HTM theory are still being developed. Today the HTM code in NuPIC can be used to analyze streaming data. It learns the time-based patterns in data, predicts future values, and detects anomalies. NuPIC includes discussion groups on HTM theory, research on extending HTM, and source code for complete applications based on HTM. We welcome participation in all of these areas.
Download, install, and run NuPIC for the first time
Please consider that NuPIC is based on principles that are unfamiliar to most people, including Hierarchical Temporal Memory (HTM), Sparse Distributed Representations (SDRs), online learning, and distributed memory. There is a steeper-than-normal learning curve.
Learn how to use NuPIC by reading our documentation below
From beginner instructions to advanced expert API reference, we hope to help you build intelligent applications.
NuPIC is an Hierarchical Temporal Memory (HTM)
HTM is a set of algorithms which model the functionality of the neocortex in the human brain. HTM Theory is the key to unlocking Intelligent Applications and Machines.
Neuroscience and Computing
Working together towards a new era of Machine Intelligence
We want to share Machine Intelligence with everyone, and support research and further development of these algorithms. There are many directions the algorithms within NuPIC could take, and we encourage you to get involved. Please read below to find out how you can interact with our development team and existing community members to push this technology into the future.
Stay up-to-date with the latest on our project and community
Read our Blog, browse recent News, learn about our Events, and see mentions of NuPIC in the Media and on the web.