Warning: Use of undefined constant add_shortcode - assumed 'add_shortcode' (this will throw an Error in a future version of PHP) in /nfs/c03/h04/mnt/49321/domains/hackingtheuniverse.com/html/wp-content/plugins/stray-quotes/stray_quotes.php on line 615

Warning: Use of undefined constant MSW_WPFM_FILE - assumed 'MSW_WPFM_FILE' (this will throw an Error in a future version of PHP) in /nfs/c03/h04/mnt/49321/domains/hackingtheuniverse.com/html/wp-content/plugins/wordpress-file-monitor/wordpress-file-monitor.php on line 39
Numenta

Numenta

In 2005, Jeff Hawkins (founder of Palm Computing and Handspring) published a book titled, “On Intelligence” that described a model of the brain and human intelligence based on the neocortex. The neocortex is the most recently evolved part of the brain and the core of human conscious perception. Hawkins suggests that a key part of the analytical ability of the neocortex is the way it compares changes over time. That “temporal” component may be an important part in trying to replicate human intelligence or create “artificial intelligence”. Hawkins has started a company to develop his ideas by producing software that mimics the structure of the neocortex and the function of human intelligence.

On Intelligence by Jeff Hawkins – [amazon.com]

On Intelligence

Numenta

About Numenta
Founded in 2005, Numenta is developing a new type of computer memory system modeled after the human neocortex. The applications of this technology are broad and can be applied to solve problems in computer vision, artificial intelligence, robotics and machine learning. The Numenta technology, called Hierarchical Temporal Memory (HTM), is based on a theory of the neocortex described in Jeff Hawkins’ book entitled On Intelligence (with co-author Sandra Blakeslee).

Introduction to Numenta Technology
Numenta is creating a new computing paradigm that replicates the structure and function of the human neocortex. This Hierarchical Temporal Memory (HTM) technology has the potential to allow computers to solve problems that are currently easily solved for humans but difficult or impossible to solve for machines. Examples include a vision system that can recognize faces or a system that can recognize dangerous traffic situations.

An HTM system is not programmed in the 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 sensor input. You might say it “learns” what objects are in the world and how to recognize them.

Hierarchical Temporal Memory – Comparison with Existing Models

The purpose of this document is to compare HTMs with several existing technologies for modeling data. HTMs use a unique combination of the following ideas:
• A hierarchy in space and time to share and transfer learning
• Slowness of time, which, combined with the hierarchy, enables efficient learning of intermediate levels of the hierarchy
• Learning of causes by using time continuity and actions
• Models of attention and specific memories
• A probabilistic model specified in terms of relations between a hierarchy of causes
• Belief Propagation in the hierarchy to use temporal and spatial context for inference

Numenta Community Wiki

Featured Articles
Vision experiment: A case study for creating your own vision experiment using NuPIC 1.6
Images Application: NuPIC 1.4 includes Images, a new application for training HTMs on visual data.
Installation and Testing: A summary of platforms that have been tested with NuPIC.
Pictures: A discussion of the Pictures example, found in the NuPIC release.
Pictures Case Study: A case study on tuning a complex application.
Script Repository: Short scripts that demonstrate NuPIC Tools for very specific functions.
Wall Street: Using the stock market as a guide, this example focuses on visualizing and understanding what is learned in the nodes of an HTM network.
Confusion Matrix Tool: NetConfusion is a tool to help you analyze the results of your experiments.
Network Topology: A discussion about determining and creating the correct topology for your network.
Using iPython: A favorite among our engineers, iPython is a useful tool for helping you debug your network.

Leave a Reply

You must be logged in to post a comment.