Imagination Engines, Inc., Home of the Creativity Machine


The Big Bang of Machine Intelligence!

Imagination Engines, Inc., Home of the Creativity Machine
The simple
  • Three Generations of Creativity Machines

    The simple, elegant, and inevitable path to human level machine intelligence and beyond, the Creativity Machine Paradigm, US Patent 5,659,666 and all subsequent foreign and divisional filings.


Wall Street Journal: Can an AI System Be Given a Patent?

Fast Company: Can a robot be an inventor?

BBC: AI system 'should be recognised as inventor'

Financial Times: Patent agencies challenged to accept AI inventor

Futurism: Scientists are trying to list AI as the inventor on a new patent

The Disruption Lab: The disruption that is DABUS: Beyond AI

ACT-IAC: The dawn of conscious computing

WIRED: This artificial intelligence is designed to be mentally unstable


IEI's Advanced Data Mining Technology

Summary - IEI has led the pack in developing advanced data mining methodologies for the interrogation of trained neural network models to elucidate the rules that have been captured therein. In addition to holding the key patents in the area of rule-extraction from artificial neural networks, we have also built free-standing products that perform these functions automatically. These software packages are either made available to those holding valid licenses to the patented IEI neural network paradigms or to those carrying out development efforts in collaborative efforts with the company. Furthermore, we have also produced an exciting form of virtual life called a "DataBot," that emulates many of the cognitive skills used by human brains in seeking and analyzing data contained within extensive databases.

Details - Having formed an accurate neural network model, there is undoubtedly some underlying logic that has been captured that can potentially reveal the key factors and relationships therein. Assisting us in these efforts is our in-house neural network trainer, PatternMaster™, that allows for rapid point and click skeletonization of trained models in 3D virtual reality so as to reveal key factors and relationships. For instance, in the figure to right, we have trained a concrete formulation network on approximately one thousand data patterns over just 1 second using our STANNO paradigm as the underlying neural engine. We have then "surgically pruned" the net (bottom view) in virtual reality to reveal critical factors and schema inherent to the concrete curing process. Apart from various aggregate concentrations, it appears that curing time (age), air content, and water content play the most important roles in determining the tensile strength of the resulting concrete formulation. More detailed analysis of the red and blue connection weights indicates that cure time positively influences concrete tensile strength, while both air and water content is detrimental to this model output factor.

More advanced versions of PatternMaster are capable of automatically identifying the underlying logic captured within the network, effectively labeling sets of connection weights producing, for instance, OR, AND, XOR, additive, multiplicative, or fuzzy logic of various kinds.

PatternMaster represents a "do-it-yourself" philosophy that IEI brought to the scene in the mid 1990s in response to those charging an arm and a leg for building artificial neural network models. Using this product, we and our joint venture partners are able to train extremely large neural network models in seconds, oftentimes enabling genomic scale analyses. The model is interrogatable even during training and is easily analyzed via our point and click methodologies in 3D virtual reality.

In addition to our rule extraction techniques, we have also developed an exciting new form of virtual life called a "DataBot." Prototyped in Microsoft Excel, spreadsheet cells, tantamount to brain cells, knit themselves together to form brain-like pathways that allow this distributed, biomimetic algorithm to follow "gradients" in its data environment toward targeted information, propelled through the worksheet by its own volition. As it mobilizes itself through the data environment, it learns without recourse to human conceived training algorithms, effectively enlisting delegated neurons to train others.


New Page 1

© 1997-2020, Imagination Engines, Inc. | Creativity Machine®, Imagination Engines®, Imagitron®, and DataBots® are registered trademarks of Imagination Engines, Inc.

1550 Wall Street, Ste. 300, St. Charles, MO 63303 • (636) 724-9000