中文 English

Architectural Considerations For AI


Custom chips, labeled as artificial intelligence (AI) or machine learning (ML), are appearing on a weekly basis, each claiming to be 10X faster than existing devices or consume 1/10 the power. Whether that is enough to dethrone existing architectures, such as GPUs and FPGAs, or whether they will survive alongside those architectures isn't clear yet. The problem, or the opportunity, is that t... » read more

11 Ways To Reduce AI Energy Consumption


As the machine-learning industry evolves, the focus has expanded from merely solving the problem to solving the problem better. “Better” often has meant accuracy or speed, but as data-center energy budgets explode and machine learning moves to the edge, energy consumption has taken its place alongside accuracy and speed as a critical issue. There are a number of approaches to neural netw... » read more

Putting Limits On What AI Systems Can Do


New techniques and approaches are starting to be applied to AI and machine learning to ensure they function within acceptable parameters, only doing what they're supposed to do. Getting AI/ML/DL systems to work has been one of the biggest leaps in technology in recent years, but understanding how to control and optimize them as they adapt isn't nearly as far along. These systems are generall... » read more

Mythic Case Study


Mythic, the provider of a unique AI compute platform, was designing an innovative intelligence processing unit (IPU) and found themselves in need of a small, power-efficient, yet programmable core to take care of specific supporting functions. As no off-the-shelf core would exactly match the needs and customization proved challenging, Mythic eventually opted for a complete solution by Codasip. ... » read more

Week In Review: Auto, Security, Pervasive Computing


Pervasive computing — Data centers, cloud, 5G, edge In a move to improve data collection for IC manufacturing, PDF Solutions entered a definitive agreement to acquire Cimetrix Incorporated. Cimetrix makes connectivity products for smart manufacturing, which PDF Solutions will use in its Exensio product to facilitate moving IC manufacturing data from the factory floor to cloud-based analytics... » read more

Week In Review: Auto, Security, Pervasive Computing


AI/Edge Arm putting AI (artificial intelligence) and machine learning (ML) on the Cortex-M processor by offering IP for a microNPU for Cortex-M. The company says in a press release that it will deliver a 480x uplift in ML performance. The new Cortex-M IP is Arm Ethos-U55 NPU, which Arm says is the industry’s first microNPU (neural processing unit). Arm is hoping the new IP will start an expl... » read more

June’19 Startup Funding


During the month of June, there were 15 startups that brought in funding rounds of $100 million or more, as investors continued to chase deals in cybersecurity, automotive technology, semiconductors, and a variety of services. There were no billion-dollar deals as spring slid into summer; yet, those 15 companies together raised a total of about $3.13 billion. Aurora Innovation, the developer... » read more

How To Manage DFT For AI Chips


Semiconductor companies are racing to develop AI-specific chips to meet the rapidly growing compute requirements for artificial intelligence (AI) systems. AI chips from companies like Graphcore and Mythic are ASICs based on the novel, massively parallel architectures that maximize data processing capabilities for AI workloads. Others, like Intel, Nvidia, and AMD, are optimizing existing archite... » read more

Using Analog For AI


If the only tool you have is a hammer, everything looks like a nail. But development of artificial intelligence (AI) applications and the compute platforms for them may be overlooking an alternative technology—analog. The semiconductor industry has a firm understanding of digital electronics and has been very successful making it scale. It is predictable, has good yield, and while every de... » read more