Deploying adaptive AI in distributed water plants - Barbara Accoina

Aspinity appoints new chief executive officer

Categories Edge Computing News  |  People
Aspinity appoints new chief executive officer

Aspinity, a company focused on analog machine learning processors, has named Richard Hegberg as its new chief executive officer. With over 25 years of leadership experience at prominent semiconductor firms such as SanDisk, Qualcomm, AMD, and ATI Technologies, Hegberg’s appointment is seen as a strategic move by Aspinity.

The company aims to leverage Hegberg’s expertise in the semiconductor industry to guide them through a critical growth phase. Aspinity believes that his leadership will enable the company to navigate the market for low-power, always-on processing technologies, ensuring competitiveness in this dynamic technological landscape.

“Richard Hegberg’s track record in elevating companies with unique breakthrough technologies to billion-dollar revenues was a decisive factor in his selection,” says Tom Doyle. “With his leadership, Aspinity is poised for significant growth in product development and IP, cementing its position as a leader in all-analog ML/AI processing.”

To address the challenge of power consumption in edge applications, Aspinity has developed a reconfigurable analog modular processor (RAMP) that brings the performance of digital neural processing systems into the analog domain. Aspinity claims that its technology achieves near-zero power consumption during inference, which is advantageous for always-on applications.

Inside the reconfigurable analog modular processor are multiple parallel, independent analog circuit blocks that operate in the subthreshold domain, functioning at voltage levels below the threshold voltage at which a transistor turns on. This design choice contributes to the processor’s energy efficiency, and each of these analog circuit blocks is intentionally compact to reduce physical space requirements.

According to Aspinity, the analog processor maintains high precision through its 10-bit analog non-volatile memory, a patented technology that’s implemented in CMOS without the need for additional components. This memory serves as storage for crucial neural network parameters such as weights, biases, and activations, significantly enhancing the accuracy of the machine learning models.

In terms of software programmability, Aspinity’s reconfigurable analog modular processor chip allows users to adjust circuit block connections and parameters. The software-based configuration can be stored in on-chip memory, further improving the flexibility and adaptability of the platform.

“It has become clear to me that Aspinity’s technology and IP portfolio deliver unique sensing capabilities at near-zero power, utilizing the analog ML processor. The company has demonstrated this value in a variety of proof of concepts across automotive, dash cam, IOT, security, and data center applications,” says Richard Hegberg.

Read more:

Seco, Qualcomm unite to advance IIoT edge computing products

Premio adds new edge AI processor to its portfolio for industrial computers

Article Topics

 |   |   |   | 


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Featured Edge Computing Company

Edge Ecosystem Videos

Machine learning at the Edge


Latest News