Using AI With Sensor And Signal Data At The Edge And In The Cloud (Internet of Things Applications USA 2017)

Sensors are everywhere - wearables, connected devices, even "old industry" products are now becoming instrumented and turning "smart." But working with sensor data - particularly sensor data in the form of signals - can be very difficult. High sample-rate signals like sound, vibration, accelerometry and electrical signals are usually the domain of the signal processing engineer, but there are new machine learning and AI-driven techniques that make these types of data much easier to work with, and much easier to blend with data from other sensors. This session will help you better understand:

 

· The difference between time-series and signal data in sensors.

 

· How to determine the right approach for working with each.

 

· How to use the latest machine learning and AI techniques on this kind of data.

 

· What's different if you need an embedded solution, and what tools are available to help.

Stuart is co-founder and CEO of Reality AI, an artificial intelligence company focusing on sensors and signals.

Stuart was previously co-CEO of LaCrosse, a technology and operations platform for investment funds that was acquired by Wells Fargo.

He has a PhD from UC Berkeley and a BA from the University of Chicago.

Reality AI enables devices to understand reality in image, sound, acceleration, vibration and other sensor inputs. Our technology is based on patented, advanced artificial intelligence techniques - optimized for sensor inputs - that get results often unattainable with other tools.