BrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, ADR: BCHPY), the world’s first commercial producer of ultra-low power, fully digital, event-based, neuromorphic AI, today released a white paper detailing the company’s TENNs-PLEIADES (PoLynomial Expansion In Adaptive Distributed Event-based Systems), a method of parameterization of temporal kernels that reduces memory and computational costs while effectively capturing long-range temporal correlations, enhancing performance across various spatiotemporal tasks.“Building Temporal Kernels with Orthogonal Polynomials” is the latest addition to BrainChip’s growing resource library that highlights the processes and technologies that make the company a leader in the neuromorphic computing space. The white paper covers topics related to temporal network architectures, event-based benchmark experiments and possibilities that such an approach can achieve.
TENNs-PLEIADES excels in directly performing spatio-temporal classification on data from event-based sensors. These sensors offer ultra-low latency, and until now, processing their signals in a principled manner has been a significant challenge. By utilizing a unique kernel parameterization based on orthogonal polynomials, TENNs-PLEIADES achieves state-of-the-art performance across three distinct spatio-temporal datasets.TENNs-PLEIADES is the latest technological advancement added to BrainChip’s IP portfolio and an expansion of Temporal Event-Based Neural Nets (TENNs), the company’s approach to streaming and sequential data. It complements the company’s neural processor, Akida IP, an event-based technology that is inherently lower power when compared to conventional neural network accelerators. Lower power affords greater scalability and lower operational costs. BrainChip’s Akida™ supports incremental learning and high-speed inference in a wide variety of use cases. Among the markets that BrainChip’s Essential AI technology will impact are the next generation of smart cars, smart homes of today and tomorrow, and industrial IoT.To access additional learning resources regarding neuromorphic AI, essential AI, TENNs and other technical achievements, interested parties are invited to visit BrainChip’s white paper library at https://brainchip.com/white-papers-case-studies/.
Expand
(20min delay)