KnowmTech, LLC Announces Recent Patent Issuances for Feature Extraction and Adaptive Control Systems

KnowmTech, LLC Announces Recent Patent Issuances for Feature Extraction and Adaptive Control Systems

Tuesday, 10 February 2015 06:00 PM

Albuquerque, NM / ACCESSWIRE / February 10, 2015 / KnowmTech, LLC announces the issuance of two recent U.S. patents:

U.S. Patent No. 8,918,353 entitled "Methods and Systems for Feature Extraction" issued on December 23, 2014 and covers a method and system for extracting features utilizing AHaH (Anti-Hebbian and Hebbian) learning. A sparse input data stream is presented to a synaptic matrix of a collection of AHaH nodes associated with an AHaH module. The AHaH module operates an AHaH plasticity rule via an evaluate phase and a feedback phase cycle. A bias input line can be modulated such that a bias weight do not receive a Hebbian portion of the weight update during the feedback phase in order to prevent occupation of a null state. The input space can be bifurcated when the AHaH nodes fall randomly into an attractor state. The output of the AHaH module that forms a stable bit pattern can then be provided as an input to a content addressable memory for generating a maximally efficient binary label.

U.S. Patent No. 8,909,580 entitled "Methods and Systems for Thermodynamic Evolution" issued on December 9, 2014. The '580 patent relates to an adaptive control system constructed based on the property of volatile matter to self-organize to maximize the dissipation of energy. The logical state of sensory nodes in a node circuit are set and projected into a network. The system evaluates the logical states of processing nodes by summing input currents of processing nodes and projecting the processing node's state into the network. The strength of the processing node is increased such that the logical state of the sensory node matches with logical states of processing nodes utilizing a plasticity rule. The system is configured to maximize energy dissipation by creating weight structures to stabilize nodes with the logical state. The internal positive feedback of the node circuit forces competition between nodes such that one node is driven to a high logical state and other nodes to a low logical state. 

About KnowmTech, LLC

KnowmTech, LLC was founded in 2002 to develop breakthrough machine learning technologies that address fundamental problems inherent in modern computing architectures and centered around the Knowm concept. The separation of memory and processing in computing architectures leads to wasted energy shuttling information back and forth. This causes tremendous problems for simulations of adaptive operations-exactly the operations needed for things like machine learning (ML). Since ML is inherently about adaptation, it turns out that building large-scale adaptive systems like brains-at a power efficiency close to biology-is simply not feasible with traditional computing approaches. 

AHaH computing changes this by taking advantage of a universal adaptive building block found throughout nature and mapping this to electronic circuits. The physical circuit turns out to be exceedingly simple, allowing for various levels of simulation abstractions-including a very efficient digital emulator that runs on modern hardware.

Contact:

Alex Nugent
[email protected]
505-988-7016

 

SOURCE: KnowmTech, LLC