Download the Knowm white paper!

By popular demand, we’ve condensed the most important general bits of knowm.org into one compact and printable document: Knowm_Whitepaper.pdf

History

The best way to predict the future is to invent it! Knowm exists to fill a niche in the evolving technological landscape and lead the computing industry to its first continuously-adaptive neuromorphic processor based on the self-organizing principles of Nature. Knowm has been growing since 2002, when lead inventor Alex Nugent began patenting his ideas around adaptive computing architectures and founded an intellectual property holding company called KnowmTech. Initial seed funding for the endeavor was made possible by business woman and entrepreneur, Hillary Riggs. The portfolio now includes over 40 patents spanning memristive components and circuits all the way to large scale neuromorphic architectures. Alex Nugent co-created and advised the DARPA SyNAPSE program and more recently has been awarded government SBIR and STTR contracts to further develop the technology.

“Perfection is Achieved Not When There Is Nothing More to Add, But When There Is Nothing Left to Take Away” –Antoine de Saint-Exupéry

In 2012, physicist, electrical engineer and software developer, Tim Molter, joined the effort to lead software development and further design chip architectures with Alex. This collaborative effort lead to the publication of the formal introduction to AHaH computing in early 2014: AHaH Computing–From Metastable Switches to Attractors to Machine Learning. More recently, Knowm has partnered with key experts in the field to further ramp up efforts on a path to commercialization. Collaborators include memristor fabrication pioneer Kris Campbell, Ph.D. from Boise State University and memristor circuit designer pioneer Dhireesha Kudithipudi, Ph.D. from Rochester Institute of Technology. Most recently, investor and consultant Sam Barakat has joined the team to help launch Knowm Inc.

AHaH Attractors

AHaH Attractors

Technology

Alex’s original idea and inspiration was “reevaluate our preconceptions of how computing works and build a new type of processor that physically adapts and learns on its own”. By observing Nature he noticed one major difference between modern computers and Nature’s brains: brains don’t separate processor from memory, as we do with a CPU and RAM. In a nervous system (and all other natural systems), the processor and memory are the same machinery. The distance between processor and memory is zero. Another observation was that whereas modern chips must maintain absolute control over internal states (ones and zeros), Nature’s computers are volatile – the components are analog, their states decay, and they heal and build themselves continuously. Pursuing this observation, Alex discovered something that was in plain sight but seldom recognized as significant: Nature’s transistor – two energy dissipating pathways competing for conduction resources. This simple adaptive building-block leads to the formation of energy-dissipating fractals found at all scales of Nature and life, from rivers to trees to lightening and brains. Driven by the second law of thermodynamics, matter spontaneously configures itself to dissipate energy! The challenge remaining was to understand how to recreate this phenomenon on a chip and interface with existing hardware to solve real-world machine computational problems. Around the same time, independent researcher Dr. Kris Campbell was perfecting ‘variable resistor’ devices, and HP made the connection to Leo Chua’s theoretical prediction of a missing circuit element he called the memristor. In part due to the influence of the DARPA SyNAPSE program, memristors began to appear in the literature. However, a unified theory on how to use these new devices in learning systems was not yet available. Thanks in part to funding from the Air Force Research Labs, Alex and Tim were able to publish the theory of AHaH Computing. Years of work designing various chip architectures and validating capabilities lead to the specification of Thermodynamic RAM or kT-RAM for short, a simple co-processor ‘core’ that can be plugged into existing hardware platforms to accelerate synaptic integration and adaptation operations. Validated capabilities of kT-RAM include unsupervised clustering, supervised and unsupervised classification, complex signal prediction, anomaly detection, unsupervised robotic actuation and combinatorial optimization of procedures – all key capabilities of biological nervous systems and modern machine learning algorithms with real world application.

kT-RAM

kT-RAM

With the arrival of physical devices capable of Back-End-of-Line integration with CMOS, a theory on how to build learning circuits, and the specification of a general-purpose learning co-processor, Knowm Inc was formed. We are now entering a phase of first-generation chip manufacturing and application development.

Mission Statement

The blueprints for Thermodynamic RAM were developed by iteratively reducing complexity at every step of the way, brutally and tirelessly stripping away layers down to the bare essentials, culminating in what we believe is a simple and elegant neuromorphic processor architecture that can be built with current technology.

Here are some of Knowm Inc’s core values, which emerged from our rigorous methodology and successful collaborations:

  1. We dig deep to understand and then creatively and purposely solve problems, brutally seeking simpler solutions.
  2. We strive to not reinvent any wheels. We leverage existing alternatives so we can focus our energy into advancing technology.
  3. We favor honesty and candor. We recognize that a mistake is the first step in learning, and we are quick to admit and fix mistakes.
  4. We achieve excellence through rigorous quantitative benchmarking and measurement. We show, don’t tell, and we believe in our work.
  5. In the face of obstacles we don’t obfuscate: we clarify.
  6. We respect the customer and work hard to understand their needs and to communicate realistic solutions.
  7. We support team member excellence and happiness and believe that the whole is greater than the sum of the parts.
  8. We are competitive. We play fair, play hard, and play to win.
  9. We respect Nature, and believe it to be the highest form of technology.
  10. We love what we do.

Founders

  • Alex Nugent

    Chief Executive Officer

    Chief Executive Officer and lead inventor of AHaH Computing and Thermodynamic RAM.

    • Tim Molter

      Chief Technical Officer

      Chief Technical Officer providing leadership and direction in all technical matters.

      • Hillary Riggs

        Angel Investor and CAO

        Early angel investor and co-founder of KnowmTech intellectual property, CAO of Knowm Inc.

        • Sam Barakat

          Investor and Consultant

          Strategic financial and economic advisor and early investor in Knowm Inc.

          University Collaborators

          • Kris Campbell

            Memristor Pioneer and Expert

            Early memristive device pioneer and BEOL CMOS memristor device fabrication.

            • Dhireesha Kudithipudi

              Biologically Inspired Computing Architectures

              Assistant Professor, Department of Computer Engineering, Rochester Institute of Technology.

              9 Comments

                • Luis Ortiz
                  reply

                  Nice website.

                  • Alex Nugent
                    Alex Nugent
                    reply

                    Thanks! Its taken a lot of work.

                • Jose
                  reply

                  As a layman with a deep interest in the coming AI era, my strong wish is that I had something of value to contribute to Knowm. Knowm’s seems by far the most holistic of approaches to this new paradigm. However, nothing impresses me more here than #9 in your Mission Statement. Take good care, and godspeed on your noble journey!

                  • Alex Nugent
                    Alex Nugent
                    reply

                    Jose-thank you for the comment. Everybody has something valuable to contribute–it usually just takes getting to know each other. Probably the most valuable thing you could do right now is to go into nature and really look at it. Observe how it is built and how it builds itself. You are not looking for complexity–there is a simplicity under it all and everybody can see it with time. Once you see it–help somebody else to see it.

                • Jose
                  reply

                  Indeed! As a result of your eye-opening video lecture, I have been observing the rich leafage in our tropical climate with a relative open mind, and it all makes such good sense… Nature’s amazing diversity, self-repairing resilience, adaptive evolution, and apparent intelligence, may lead many to the conclusion that there must be a super-intelligent deity behind it all. But as you’ve noted so well, it all points to the physical/thermodynamics laws in our universe leading to an Occam’s razor in energy containment and dissipation.
                  Thank you for all your time and effort spent to present these eye-opening ideas in such clear fashion. Looking forward to living long enough to see us outsmart 4 billion years of natural evolution! 🙂

                • Roland d'Humieres
                  reply

                  Amazing job! Amazing Team! Amazing Core Values! Amazing everything!
                  Congratulations!

                • ed fernandez
                  reply

                  truly amazing what you are doing, congrats, wishing you all the best

                • Antoine
                  reply

                  Hi,
                  This idea of sticking to nature is amazing, and the laws you infer are inspiring. Bravo. I am discovering the world of cognitive computing, and I wonder how far hardware has grown up today.
                  What are the two integrated devices that you are looking at in the box on the team photo ?

                  • Tim Molter
                    Tim Molter
                    reply

                    @Antoine That’s two memristor chips from our first development batch.

              Leave a Comment