Imagination Engines, Inc., Home of the Creativity Machine


The Big Bang of Machine Intelligence!

Imagination Engines, Inc., Home of the Creativity Machine
The simple
  • IEI Patent Overview

    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.


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


The Ultimate Idea - The Creativity Machine Paradigm
What is the ultimate idea?

We, along with many others, believe that grandest of all notions is how ideas themselves are formed within the biological neural networks of the brain. If the principle behind this highly prized cognitive mechanism can be captured, understood, and then implemented within lightning fast machine intelligence then we have attained the ultimate idea, the one that can generate all subsequent art, inventions, and discoveries.

In 1994, this company achieved that goal, announcing the accomplishment through a patent called the "Creativity Machine," a computational paradigm, already 20 years in the making, that came the closest yet to emulating the fundamental neurobiological mechanisms responsible for idea formation. Appropriately, the resulting patent's abstract (US 5,659,666) reads as follows:

A device for simulating human creativity employing a neural network trained to produce input-output maps within some predetermined knowledge domain, an apparatus for subjecting the neural network to perturbations that produce changes in the predetermined knowledge domain, the neural network having an optional output for feeding the outputs of the neural network to a second neural network that evaluates and selects outputs based on training within the second neural network. The device may also include a reciprocal feedback connection from the output of the second neural network to the first neural network to further influence and change what takes place in the aforesaid neural network.

Within months of this patent's issue, international corporations, the US government, and the press (e.g., MSNScientific AmericanNew Scientist, and Economist, among others) were asking for more information, leading to diverse contracts in materials discovery and consumer products. In 1997 this same sentient neural paradigm was chosen by a major US defense contractor as the control system behind a constellation of communication satellites that contemplated how best to distribute petabytes of critical information to the country's armed forces around the world. It worked by imagining various scenarios and then solving for the best scheduling and satellite attittude adjustments to deliver the time-critical packets. In short, the neural algorithm worked both with and against itself to optimize bandwidth in exercises vastly more important than any board game.

It was at this point that this neural paradigm inspired, and some cases, produced derivative neural network technology, leading to even more, fundamental artificial intelligence patents. Chief among these new concepts were artificial neural network objects that trained without recourse to the standard, human-conceived learning algorithms. Instantiated with billions of connection weights, they could learn from live, video streams, using the personal computer technology of 1998! Soon systems of these so-called STANNOs could self-organize themselves into vast brain-like structures called SuperNets. rather than be manually arranged in successive layers by human software engineers, as in deep learning schemes. It was these networks of networks that paved the way to such innovative automotive machine vision applications that we all know and love, such as lane departure warning, drowsiness detection, automatic high-beam control, and pedestrian/vehicle detection. Then, these compound nets could be stimulated, just as in the fundamental Creativity Machine architecture, to produce even more complex ideas, and in the case of battlefield robots and space exploration platforms, improvise clever strategies to autonomously achieve their goals and deal with the unexpected.

Then, in 2005, Creativity Machines became even more powerful, by allowing critic nets to work cooperatively or adversely in reinforcing the very best of its self-originated ideas within itself. The resulting patent's abstract (US 7,454,388) reads as follows:

A discovery system employing a neural network, training within this system, that is stimulated to generate novel output patterns through various forms of perturbation applied to it, a critic neural network likewise capable of training in situ within this system, that learns to associate such novel patterns with their utility or value while triggering reinforcement learning of the more useful or valuable of these patterns within the former net. The device is capable of bootstrapping itself to progressively higher levels of adaptive or creative competence, starting from no learning whatsoever, through cumulative cycles of experimentation and learning. Optional feedback mechanisms between the latter and former self-learning artificial neural networks are used to accelerate the convergence of this system toward useful concepts or plans of action.

In addition to producing whole new conceptual discoveries, such brainstorming neural nets became the creative AI behind real and virtual robots that could develop their behaviors from scratch and then improvise in responses to unforeseen circumstances. Such improvisational control systems greatly contributed to flight robotics at NASA, allowing space vehicles to autonomously rendezvous and dock, as well as enable the fully autonomous operation of off-word robots.

Soon, IEI will unveil its most advanced technology yet, codenamed "Sunshine Transit," a connectionist methodology that vastly exceeds so-called "deep learning" techniques in its pattern recognition capabilities, while enabling idea formation on a scale that can't be touched by modern day supercomputer technology. Arguably the first sentient machine intelligence, it can select those notions and strategies it feels are the most appropriate for a given situation or challenge in terms of its imagined consequences.

And now, this same technology, going by the name "Device for the Autonomous Bootstrapping of Unified Sentience" (DABUS), has achieved some very profound results, developing whole new and patentable concepts. No, this is not a parametric optimization with novel twists, as in early Creativity Machines or genetic programming. This amazing system is originating new ideas without any clear objectives. The controversial results have become the subject of patent applications filed around the world, prompting legal authorities to consider whether machines can own their own intellectual property (e.g., BBC News Futurism, Fast Company, and ABA).

DABUS, is about to make good the promise of a well-known NASA visionary, Dennis Bushnell, that the Creativity Machine is AI's best bet at creating human to trans-human machine intelligence. That this fundamental AI principle could be applied across so many different problem areas such as theoretical chemistry, art, music, law, machine vision, medicine, and robotics, serves as testament to the power and flexibility of this all neural, general problem solving methodology that this company invented and memorialized through dozens of U.S. and international patents, as well as a long series of peer-reviewed academic papers. 

Recommended Reading

An early explanation of the Creativity Machine - Neural Nets That Autonomously Create and Discover, PC AI , May/June, 16-21, 1996.

Using the Creativity Machine to defend Alan Turing's thesis that machines may become both intelligent and conscious -The Creativity Machine Paradigm: Withstanding the Argument from Consciousness, APA Newsletters, Volume 11, Number 2, Spring 2012.

Basic theory and history of the Creativity Machine - The Creativity Machine Paradigm, Encyclopedia of Creativity, Invention, Innovation, and Entrepreneurship, (ed.) E.G. Carayannis, Springer Science+Business Media, LLC, 2013.

Basic theory and history of the Creativity Machine - The Creativity Machine Paradigm, Encyclopedia of Creativity, Invention, Innovation, and Entrepreneurship, (ed.) E.G. Carayannis, Springer Science+Business Media, LLC, 2017.

A comprehensive theory of consciousness based upon the Creativity Machine - A Synaptic Perturbation and Consciousness, International Journal of Machine Consciousness, Vol. 06, No. 02, pp. 75-107, 2014.

How Creativity Machines quantitatively predict the tempo with which we think and their distinct advantage in building synthetic brains - Pattern Turnover within Synaptically Perturbed Neural Systems, Procedia Computer Science, 88, Elsevier, 2016.

The inevitable tradeoff between creative brilliance and cognitive malfunctions in both Creativity Machine-based synthetic and human intelligence - Cycles of Insanity and Creativity within Contemplative Neural Systems, Medical Hypotheses, 94:138-147, Elsevier, 2016.

A layman's summary of how Creativity Machines reveal help us to understand the link between creative brilliance and psychopathologies - A neurodynamic model linking creativity and insanity, Atlas of Science, 2017.

New Page 1

© 1997-2019, 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