A Uniform, Tessellated Architecture for Energy-Efficient Learning and Inference

Jerry Felix, Steve Brunker, Carol Hibbard

Presentation: In-Person

Keywords: energy-efficient AI, tessellated architecture, cellular automata, wave-based learning, distributed intelligence, collision-based memory, estimator, edge AI, low-power inference, spatial computing, Estimator, edge computing

Presenter Full Name: Jerry Felix

TL;DR: A tessellated, low-power AI architecture enables real-time learning and inference through local wave interactions and distributed memory, without centralized control or complex computation.

Presenter Email: [e-mail]

Abstract: This paper presents a novel architecture designed from first principles to support real-time learning and inference through the interaction of energy-efficient components. Each element operates autonomously, storing knowledge, reacting to inputs, and forming predictions without centralized control, complex arithmetic, or external scheduling. Inspired by biological systems and implemented as a uniform grid of identical cells, the architecture enables scalable, distributed intelligence using wave-based communication, collision-driven memory, and geometry-based predictions. We describe the logical and physical layers, detail subsystem roles, and explain how small components, tessellated across space, can deliver sophisticated behavior, suggesting a shift in AI design.

Presenter Bio: Jerry Felix is the Chief Architect and co-founder of Brain-CA Technologies, a Cincinnati-based AI startup pioneering energy-efficient, first-principles AI architectures. With over 40 years in IT, including 15 years at Hewlett-Packard, he has patented a novel AI system that reduces training time, cost, and energy consumption. Felix is also the author of The Intelligence Shift, which outlines Brain-CA’s groundbreaking approach to AI design.

Paper Checklist Guidelines: I certify that all co-authors have validated the presented results and conclusions, and have read and commit to adhering to the Paper Checklist Guidelines, Call for Papers and Publication Ethics.

YouTube Link: https://youtu.be/3dNA3hZkgqI

YouTube Link Poster: https://youtu.be/2uF_3bkgPlE

Dataset Release: I certify that all co-authors commit to release the dataset and necessary scripts to reproduce the presented results.

Google Slides: n/a - poster presentation only.

Poster: Yes

Workshop Registration: Yes, the presenter has registered for the workshop.

YouTube Link Short: (coming soon)

Submission Number: 5