Pattern Computer Releases Seminal Paper on the foundations of Explainable AI

REDMOND, WA, December 9, 2021 – Pattern Computer,® Inc. (PCI) has posted a paper on ARXIV.org1 that providing a detailed description of how explainable AI will modify the scientific method, and open new avenues for scientific research and commercial applications.

The paper suggests that today’s AI applications are dominated by neural network models. While progress is being made in achieving high predictive success using these models, most internal processes remain “black boxes,” preventing comprehension of the logic and processes executed within the model’s interior layers. This opacity limits usefulness in many applications, including in regulated industries requiring such knowledge. Examples include the GDPR in the EU, and autonomous car and plane certifications.

From the paper:

“…(W)e outline emerging opportunities and challenges to enhance the utility of AI for scientific discovery. The distinct goals of AI for industry versus the goals of AI for science create tension between identifying patterns in data versus discovering patterns in the world from data. If we address the fundamental challenges associated with “bridging the gap” between domain-driven scientific models and data-driven AI learning machines, then we expect that these AI models can transform hypothesis generation, scientific discovery, and the scientific process itself.”

The authors recognize a data future driven by hyper-growth in parameterization, ever-higher dimensional data reduction requirements, and increasing complexity of mixed data sets.

From the paper:

“To deliver on this promise, however, the components, principles, and interactions captured by these AI models, and how they interact with broader domain-specific methodologies and constraints, must be understood better. These models currently remain largely opaque, in part since they are remarkably complex, but also since they have thus far been driven by generalization and industrial goals, rather than extrapolation and scientific goals. The capacity to understand data representations and principles learned by AI models constitutes a tantalizing prospect—and grand challenge—with the potential for significant impacts on scientific discovery and the origination of novel theory.”

“At Pattern Computer, we believe that XAI will bring with it a pivotal moment both in the practice of science, and in the solution of business problems. Until now, AI – and the scientific method – have required the user bring an initial hypothesis to the problem. But requiring an intuitive hunch at the start is, in our view, a complete fail. Now, for the first time, we can start with the data, ask XAI to help us see how it made predictions, and in so doing get a deep understanding beyond our original knowledge. This is more than exciting,” Mark R. Anderson, CEO of Pattern Computer, added.

The Pattern team believes that transparency of models, and the inclusion of increasing amounts and types of relevant data, will brings insights, discoveries and breakthroughs in all industries. Drug discovery and design, material science, supply chain optimization, cancer detection and cures, real time diagnostic testing for COVID and other diseases, and many other applications will be driven by XAI.

We invite you to learn more by reading the full text at


About Pattern Computer

Pattern Computer, Inc., a Seattle-area startup, uses its proprietary Pattern Discovery Engine to solve the most important and most intractable problems in business and medicine. Its proprietary mathematical techniques can find complex patterns in very-high-order data that have eluded detection by much larger systems.

While the company is currently applying its computational platform to the challenging field of drug discovery, it is also making pattern discoveries for partners in several other sectors, including additional biomedical research, materials science, aerospace manufacturing, veterinary medicine, air traffic operations, and finance.


 Brad Holtz -

 301.529.9944 -

Copyright © 2021 Pattern Computer, Inc. All Rights Reserved. Pattern Computer Inc., and PCI are trademarks of Pattern Computer, Inc. or its subsidiaries. Other trademarks may be trademarks of their respective owners.

1 Learning from learning machines: a new generation of AI technology to meet the needs of science, Luca Pion-Tonachini, Kristofer Bouchard, Hector Garcia Martin, Sean Peisert, W. Bradley Holtz, et. al,