Deeplite supports tinyML Foundation as strategic partner
EP&T MagazineAutomation / Robotics Electronics Engineering Software Engineering AI deep learning optimization software
Partnership will help advance compact AI for computer vision applications on Arm devices
Deeplite, a Montreal-based provider of AI optimization software designed to make AI model inference faster, more compact and energy-efficient, announced it has sponsored the tinyML Foundation as a platinum partner and a member of its strategic partner advisory board. Deeplite is a charter member of regional tinyML Meetups in Canada through its local chapter in Montreal, as well as Toronto.
With over 10,000 members, the tinyML Foundation plays an important role in growing the ecosystem to support the development and deployment of ultra-low power machine learning solutions at the edge, setting its technical direction and roadmap, promoting sustainability and educating key audiences. Deeplite is joining a panel of over thirty organizations, including Qualcomm, Samsung Electronics, Arm AI, and NXP Semiconductors. In addition, several academic institutions are active in the organization, including Harvard University, Stanford University, and edX. Members gather at global quarterly events held in different regions, including their flagship tinyML Summit event (March 28-30, 2022 in Burlingame, CA).
Machine Learning (ML) is a dynamic sector that benefits products that many people use every day. However, it is challenging to deliver ML on cost-effective devices due to constrained computing resources and power limits. To enable applications closer to the user, it is critical to find ways to facilitate ML inference on small, resource-constrained devices typically “on the edge” of the Cloud. This led to the development of tiny Machine Learning or tinyML, a powerful new trend to enable cost-effective devices to use advanced ML to deliver computer vision and other perception tasks.
More edge devices in AI & deep learning strategies
“There are billions of small, connected-devices everywhere. The challenge is that they have very limited resources, so how can we add intelligence to them?” said Davis Sawyer, chief product officer and co-founder at Deeplite. “Being a Platinum Strategic Partner of the tinyML Foundation helps us advance the technology vision of tinyML and bring it to real-world applications. It is important to evangelize the efficiency benefits of tinyML and discover new use cases, while bringing together the academic, startup and commercial communities to work together more closely.”
Deeplite recently announced DeepliteRT, a new addition to its platform that accelerates AI model inference on low-power Arm CPUs for production deployments. As organizations look to include more edge devices in their AI and deep learning strategies, they are faced with the challenge of making AI models run on edge devices, including security cameras, commercial drones, and sensors that often have very limited power budgets and processor resources.
DeepliteRT solves this challenge with an innovative way to run ultra-compact quantized models on commodity Arm processors.
“We look forward to collaborating further with Deeplite as a Platinum Strategic Partner, as we work together to enable edge AI deployments for more devices and applications,” said Ira Feldman, managing director of the tinyML Foundation. “Innovative technology from Deeplite and others will play a key role in accelerating the growth of the entire tinyML ecosystem and help lead advancements on a global scale.”