Virtualization Technology News and Information
Article
RSS
AI Innovators Join Forces in Consortium for Development and Commercialization of Best-in-Class AI Computing Platform

Crossbar Inc., Gyrfalcon Technology Inc., mtes Neural Networks Corporation (mNN) and Robosensing Inc. today announced an AI consortium that will deliver a vastly accelerated, power-saving AI platform and standard that enables new AI-rich capability for edge computing, gateways, cloud and data centers.

The new group, called SCAiLE (SCalable AI for Learning at the Edge), is already working with Japanese authorities to review opportunities for the 2020 Olympics, including video-based event detection and response capability.

The organization will combine advanced acceleration hardware, resistive memory (ReRAM), optimized neural networks to create ready-made, power-efficient solutions with unsupervised learning and event recognition capability.

The consortium addresses the restrictions of traditional AI methodologies that depend on classification of data. The huge growth of IoT systems including thousands of remote edge devices such as sensor-equipped cameras creates a torrent of unstructured information in multiple forms that pours into cloud-located servers and that cannot be handled effectively by classification alone.

In contrast, SCAiLE will provide for the first time fast, scalable, and power-saving accelerators that use optimized neural networks at the edge to quickly analyze and respond to multi-modal information (including video, images, speech, keywords and sensor feeds). The platform will enable distributed AI applications rather than a centralized cloud system with high latency.

The accelerators will rely heavily on new AI accelerator architectures developed by Gyrfalcon; Crossbar's massive search across multi-modal datasets based on high bandwidth ReRAM memory; mtes Neural Networks edge and IoT device technology; and Robosensing's integration of neural network training and optimization. The organization will ensure technical interoperability and quick integration of advanced technologies from SCAiLE members.

"This is a unique opportunity to apply the benefits of ReRAM's low power and advanced performance advantages, coupled with deep neural network algorithms, to accelerate machine learning and create new exciting applications in AI," said George Minassian, CEO of Crossbar. "We have been working closely with Gyrfalcon Technology Inc., mtes Neural Networks Corporation and Robosensing, and expect rapid progress in designing the platform."

"It's a clear sign that a technology has reached its stride when customers call for inter-operability," said Jim Handy, an industry analyst at Objective Analysis. "These companies are clearly blazing a trail for others to follow as more and more systems adopt remote processing and AI to filter data so that data centers won't be overloaded with data from the edge."

"Combining the high performance with the low power of GTI's neural net accelerator with the advantages of ReRAM technology from Crossbar, along with the advanced algorithms from Robosensing and mtes Neural Networks will greatly accelerate the adoption of AI at the edge," said CEO Kimble Dong from GTI. "The results of our collaboration will simplify the development process for new products and provide a unique ability to self-learn at the edge."

"The large volume of new kinds of information cannot be handled by mere classification," said Matt Kobayashi, CEO of Robosensing Inc. "We need new ways of handling unstructured data at the edge, and the planned SCAiLE platform can help us get rid of the ‘tyranny of data classification' through power-saving self-learning devices that use clustering to detect and interpret events."

"Robosensing is looking forward to being part of SCAiLE to help lead the integration for the next generation of AI learning events. We are positive that this innovative technology will be the next chapter in AI, Learning at the EDGE, and IoT to enable learning for all types of applications."

"We continue to see a large opportunity for edge-based safety and security devices equipped with event recognition for smart cities," said Mr. Takaro Harada, CEO of mtes Neural Networks. "SCAiLE offers mNN the ability to create high-performance power-saving capability that can be incorporated into our on-site equipment for both urban and remote areas."

Published Tuesday, February 19, 2019 10:09 AM by David Marshall
Filed under:
Comments
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
Calendar
<February 2019>
SuMoTuWeThFrSa
272829303112
3456789
10111213141516
17181920212223
242526272812
3456789