BrainChip will showcase the enhancements of the second-generation Akida architecture platform to attendees at the tinyML Summit 2024, April 22-24 at the Hyatt Regency San Francisco Airport in California.
At the summit, Chris Jones, Senior Product Manager at BrainChip, will showcase Akida's enhancements, including the ability to learn effectively on extremely small form factors, as well as features such as head posture and emotion classification, all done in real-time at the edge without cloud connectivity. Company executives will be on hand to give presentations and demonstrate the features of some of the sustainable and efficient AI solutions that leverage Akida's intellectual property.
As part of the show, BrainChip will be showcasing their latest Akida technology to enable more advanced, powerful, and accurate artificial intelligence. Combining NVISO Group's AI human behavior software and BrainChip's Akida Neuromorphic, the resulting system monitors the user's state and instructs emotional reasoning by perceiving and observing head and body postures, eye tracking, and gaze in real time.
"BrainChip is a market leader in neuromorphic computing, which enables our customers to build very low-power AI devices at the edge. "Neuromorphic technology excels at the edge where most of the workloads are streaming, because each frame of data is highly correlated with the previous frame," said Tony Lewis, CTO at BrainChip. "We can use these correlations to dramatically reduce our electricity demand. In contrast, cloud-based AI has evolved a hardware ecosystem that is optimized for unrelated workloads. We feel like we're going to have an edge over companies that are just trying to enable cloud computing at the edge. At the edge, you have to rethink the hardware from the ground up. That's what BrainChip does. ”
The tinyML Summit is impactful with exhibitions, presentations, and networking. The exhibition area showcases the leaders of tinyML technology and showcases the latest innovations on site. The summit will provide multiple opportunities to network with other community members, share ideas, build collaborative solutions, and strengthen business networks.
BrainChip's neural processor, Akida IP, is an event-based technology that consumes less power than traditional neural network accelerators. Lower power consumption provides greater scalability and lower operating costs. BrainChip's Akida enables incremental learning and high-speed inference in a variety of use cases. Market segments that will be impacted by BrainChip's Essential AI technology include next-generation smart cars, smart homes now and in the future, and the Industrial Internet of Things.
BrainChip is a global leader in on-chip processing and learning for edge AI. The company's first fully digital, event-based AI processor, AkidaTM, uses neuromorphic principles to mimic the human brain, analyzing only the necessary sensor inputs at the point of acquisition to process the data with unmatched efficiency, precision, and energy economy.
Akida's unique edge learning capabilities are on-chip and cloud-independent, dramatically reducing latency while improving privacy and data security. Akida Neuroprocessor IP can be integrated into SoCs of any process technology, showing tremendous advantages in today's workloads and networks, and providing a platform for developers to create, tune, and run their models using standard AI workflows such as Tensorflow/Keras. To enable effective edge computing to be ubiquitously deployed in real-world applications such as connected cars, consumer electronics, and the Industrial Internet of Things, BrainChip is proving that on-chip AI with proximity sensors is the future of its customers' products and the planet.