tinyML Summit and Research Symposium Go to the tinyML Foundation website and register here.
Sensors and actuators with their resource-constrained processing and connectivity functions, are scaling across a wide variety of application areas: industrial, medical and healthcare, transportation, consumer and personal devices, urban management, and robotics.
Today, sensors are generating an unprecedented amount of information that – when sent to the cloud as raw data – are responsible for the explosion of energy consumption required to deal with their centralized intelligent processing. For a more sustainable future, the solution is to distribute and deploy tiny Machine Learning techniques at the sensor and actuation level.
tinyML is a fast-growing field of Machine Learning technologies and applications including tools, hardware, algorithms and software capable of performing on-device sensor data analytics at extremely low power, enabling a variety of always-on use-cases and targeting battery-operated devices.
To address the above-mentioned applications, the tinyML community, which began in 2018, now includes several hundreds of companies and universities across the world. Together we are developing Machine Learning algorithms, architectures, techniques, tools, and system approaches enabling on-device analytics for a variety of sensors (vision, audio, motion, environment, chemical, and more) at a power range below the mW level required for always-on and battery-operated devices.
Since 2019, STMicroelectronics is an active contributor to the tinyML community and continues to provide outstanding support as a Strategic Partner and Gold Sponsor. This year ST will again participate in the Summit with a keynote speech by Andrea Onetti, Executive Vice President, MEMS Sensors Sub-Group, two poster sessions, a workshop and several demonstrations of our innovative products and technologies.
Join us and contribute to the development of tiny Machine Learning!
|Plenary keynote March 29 at 15:45 PST |
Andrea Onetti: Brains into sensors with AI in the Edge
AI solutions can be made significantly more efficient when data pre-processing and initial analysis is performed as close as possible to the sensing and actuating elements, rather than in the cloud. Applying this approach vastly reduces amount of data transferred and offers enhanced data security and privacy. It decreases the processing and data storage resources required in cloud servers while allowing processing to take place in power-optimized components like ultra-low power microcontrollers and sensors. It also minimizes latency allowing real-time responses in critical situations.
Today MEMS sensors with embedded AI can operate at microwatt levels of power consumption, ultra-low latency, and minimized silicon area thanks to on-die integration of sensor and logic processing. These devices are paving the way for the Onlife Era, where embedded sensors are enabling innovative products to sense, process and take actions. To demonstrate this is already a reality today, ST will present the world’s tiniest sensor solution that is bringing intelligence in the edge.
Conference agenda – ST sessions at the SUMMIT – 29-30 March
|29 March||From 15:45||Sensing for tinyML - keynote |
Brains into sensors with AI in the Edge
|Andrea Onetti |
ST keynote speaker
|30 March||12:40-13:25||tinySW/Tools: Enabling tiny experiences for the real-world |
Ecosystem of tools for better productivity
|Danilo Pau |
|30 March||15:10-17:30||Workshop: Automated TinyML design||Danilo Pau |
|29 March |
|Poster session |
Hybrid quantization for Vocal Command Recognition on micro-controller
|Danilo Pau |
|29 March |
|Poster Session |
On-sensors AI with novel ST sensors: performance and evaluation in a real application scenario
|Michele Magno |
ETHZ University presenter
Complexity bounded classification of fisheye distorted objects with micro-controllers
|Danilo Pau |
Tiny decomposition of complex neural networks for heterogeneous micro-controllers
|Biagio Montaruli |
STMicroelectronics available demonstrations in the Exhibition Hall: