SAN JOSE, Calif., June 4, 2018 – Dr. Radoslav “Rado” Danilak, co-founder and CEO at Silicon Valley startup Tachyum Inc. will for the first time publicly talk about architecture high-performance symbolic and general AI, as well as convolution networks, at the Machine Learning & AI Developers Conference in the Santa Clara Convention Center in Santa Clara, Calif. June 6 at 2:30 p.m.
Danilak has more than 25 years of industry experience and over 100 patents designing state-of-the-art processing systems. His presentation, “Nanometer Size Semiconductors for General and Symbolic Artificial Intelligence,” will discuss how and why he believes designing around very fast transistors and very slow wires in deep nanometer fabrication nodes is required to break through the single-threaded processor performance plateau that has plagued the industry for more than a decade.
“Computationally order of magnitude efficient approaches, like symbolic and general AI require very different chips than Intel, GPUs from nVidia or TPUs from Google,” said Danilak. “The processors we build over the next 15 years will drive a trillion-dollar AI-driven IT market by 2035, so we had better get it right. Tachyum has developed a chip architecture that moves beyond the limitations of semiconductor device physics, which were thought to be insolvable, to overcome the world’s most-complex problems in data centers while also demonstrating the best AI performance anywhere.”
Tachyum’s new processor family – codenamed “Prodigy” – offers the programming ease comparable to a CPU with performance and efficiency comparable to GP-GPU, for a universal-purpose processor that can handle hyperscale workloads, AI, HPC and other demanding applications with ease. A typical hyperscale data center using servers equipped with Prodigy will provide 10 times the compute performance at the same power budget. Prodigy will reduce data center TCO by a factor of four; conversely, a Prodigy-based data center delivering the same performance as conventional servers can be built in as small as 1 percent the space and consume one-tenth the energy.
The Machine Learning & AI Developers Conference is designed specifically for developers, engineers and managers developing machine learning and artificial intelligence technologies and applications. The conference offers a unique blend of market insights, technology forecasts and practical assessments of the challenges and opportunities of machine learning and artificial intelligence in embedded systems. In addition, the Machine Learning Developers Conference features an exhibition showcasing demos and technologies focused on machine learning and artificial intelligence implementations.
Named for the Greek “tachy,” meaning speed, combined with “-um,” indicating an element, Tachyum emerged from stealth mode in 2017 to engineer disruptive intelligent information processing products. Tachyum’s founders have a track record of solving problems caused by device physics to deliver transformational products to market.