In a world in which untamed data brings expense and lost opportunity, making sense of those myriad data points is an increasingly important need but can be fraught with incredible challenges. The volume, velocity, and variety of data generated by devices operating at the edge are increasing exponentially. The potential value is huge for organizations that successfully make sense of all of this aggregated data in an actionable way while avoiding the need to leverage latency-inducing centralized cloud services.
This week, The Linley Group recognized BrainChip’s Akida Spiking-Neural-Network (SNN) processor as an enabler for edge-centric workloads. In its analysis, Linley recognized a unique Akida property – a single hardware platform that can perform as an inference engine for the Convolutional Neural Networks (CNNs) of today and support SNNs of tomorrow with its unique on-chip learning algorithms. Further, in a time in which focus on green solutions is no longer optional, Akida works its magic inside an incredibly small power budget.
• See the general architecture of Akida and discover how it’s able to work its magic.
• Discover the two methods by which the Akida technology can be acquired
• Learn what separates Akida from the competition, including its light power budget, inference + learning capability, and complete package design
While you’re here, we invite you to check out the full presentation from the Linley conference where Anil Mankar presented the Akida architecture and key features.