Today MemComputing announced that its Xtreme Performance Computing (XPC) Software as a Service (SaaS) will be made free for evaluation to all researchers and developers working in response to the COVID-19 crisis. This initiative comes from MemComputing’s ambition to identify opportunities where its technology may aid the fight against the global pandemic. “MemComputing harnesses the power of physics to dramatically reduce compute times for today’s most complex computational problems associated with optimization, big data analytics, and machine learning. By providing a free evaluation of the MemCPU XPC SaaS, MemComputing hopes to accelerate the process of finding solutions to the COVID-19 crisis on a global scale.”
Whitepaper: Accelerate Training of Deep Neural Networks with MemComputing
“The paper addresses the inherent limitations associated with today’s most popular gradient-based methods, such as Adaptive Moment Estimation (ADAM) and Stochastic Gradient Descent (SGD), which incorporate backpropagation. MemComputing’s approach instead aims towards a more global and parallelized optimization algorithm, achievable through its entirely new computing architecture.”
Podcast: The Turing Machine is Sequential, so how about a Parallel Machine?
The @RadioFreeHPC team is joined by the folks at MemComputing, a San Diego startup that’s built a new parallel foundation for computing. It calls it the universal memcomputing machine, and a realization of self-organizing circuits. “We analytically prove that the memory properties of UMMs endow them with universal computing power—they are Turing-complete.”
Chevron Technology Ventures Selects MemComputing for Prestigious Catalyst Program
Today MemComputing announced it has been selected for participation in the Chevron Technology Ventures Catalyst Program. CTV’s Catalyst Program supports early-stage companies who are advancing cutting-edge technologies that can directly benefit the oil and gas industry. “While the compute problems we will explore together are quite complex, we are currently achieving results with academic and industrial problem sets that suggest our technology is very applicable to oil and gas in areas such as demand forecasting and resource planning.”
Researchers Build Memcomputing Prototype
Over at Scientific Advances, a newly published paper describes a new high-efficiency computing paradigm called memcomputing. Modeled after the human brain, a memprocessor processes and stores information within the same units by means of their mutual interactions. Now, researchers have built a working prototype.