We held our next tinyML Talks webcast. Amit Roy from AigenEdge Private Limited presented DNN based AI application “Everywhere and Anywhere” on August 3, 2021.
DNN based AI-ML application future growth will be considerable even though the projected growth in this area has not yet taken place.
It is critical that we break through the three primary bottlenecks to promote wider DNN use in industries that are
• The DNN-based application requires a high level of skill
• The deployment cost is high.
• Unreliable/Untrustworthy DNN application giving too high a false alarm rate.
During my talk, I will go through some of the main problems and alternative ways of getting around them. DNN-based AI/ML applications will become more widespread as a result.
By putting forth a high-level perspective on it, I’ll discuss TinyML, which is geared toward being a good value in price and power consumption.
Using the hardware with little processing power and limited memory, TinyML is designed to operate on a low-powered chipset that does not require much processing power and memory.Because the computing environment was created primarily for tiny inference workloads, this is conceivable.
It has the potential to fundamentally change the Internet of Things in the future.
Dr. Amit Roy received his undergraduate degree in E&C from Delhi College of Engineering and his master’s and doctoral degrees from the University of California-Berkeley. He spent over 17 years in a chip design business before founding AigenEdge, a start-up focused on developing technologies for the tiniest, lowest power, fastest DNN with built-in explainability. He has been awarded sixteen patents by USPTO.
Watch on YouTube:
Download presentation slides:
Feel free to ask your questions on this thread and keep the conversation going!