A new associative learning-based technique could pave the way toward faster, more efficient optical computing.
For those who are at the intersection of AI hardware and software, the open source Apache TVM effort is already well known and used among a number of
IBM is reaching beyond digital technologies with a capacitor-based cross-point array for analog neural networks, exhibiting potential orders of magnitude improvements in deep learning computations. Analog computing architectures exploit the storage capability and physical attributes of certain memory devices not just to store information, but also to perform computations. This has the potential to greatly reduce the time and energy required by computers because data doesn't need to be shuttled between the memory and processor. The drawback could be a reduction in computational accuracy, but for systems that do not require high accuracy, it is the right trade-off.
Clear Ventures managed to stay out of the AI chip startup investment fray of the last several years. What it saw then was a lack of focus on software and
AI Hardware Accelerators For Machine Learning And Deep Learning. AI accelerators can significantly reduce the time to train and execute an AI model and perform specific AI tasks that cannot be conducted on a CPU.
The Verge is about technology and how it makes us feel. Founded in 2011, we offer our audience everything from breaking news to reviews to award-winning features and investigations, on our site, in video, and in podcasts.
Uppercase-tech companies are shifting their strategy away from China as their eyes turn toward Mexico for the production of critical artificial intelligence (AI) hardware components. Nvidia, Amazon, Alphabet, and Microsoft are reported to have engaged Hon Hai Precision Industry, aka Foxconn Technology, to urge expansion in production capabilities within Mexico. Further, that is seen as […]
Many artificial intelligence innovations hit a wall when it comes to applying them to real production settings.
Technology is a fast-paced field where invention has no boundaries. The world of AI-powered hardware has been fast changing since the launch of OpenAI's
In a wide-ranging conversation, Google CEO Sundar Pichai talks about the profound transformation of the company and of its core search service.
Smartphones aren't dead, they've just stagnated. Over the years, companies have tried hard to develop 'the next thing', experimenting with folding phones, AI wearable pins, and even AR/VR headsets... but here's what nobody's realized yet. There's nothing wrong with the smartphone's format. It's just lacking the next big technological leap. And that leap doesn't mean
The Texas Instruments AM69 Applications Processor features a powerful 8-core Arm Cortex A72 CPU cluster and a deep neural network accelerator with up to 32 T...
Bursting dynamics that mimic the functions of the human brain pave the way for more efficient AI systems.
RISC-V is the new hotness, and companies are churning out code and announcements, but little actual hardware. Eventually, we’re going to get to the point where RISC-V microcontrollers and SoC…
Power consumption, latency, and privacy concerns for NPUs included in application processors.
Artificial-intelligence inference workloads and the computing architectures needed to support them are rapidly evolving. Theyâre also becoming
A research team led by the University of Washington has developed an optical computing system for AI and machine learning that not only mitigates the noise inherent to optical computing but actually uses some of it as input to help enhance the creative output of the artificial neural network within the system.
Computing with light could slash the energy needs of neural networks
Xuan "Silvia" Zhang's lab at Washington University in St. Louis has reached a theoretical limit for efficiently converting analog data into digital bits in an emerging computer technology.
The board delivers 1.4 TOPS/W for drones and small robots.
