Two Los Angeles County residents face federal charges after they were arrested on suspicion of illegally exporting tens of millions of dollars' worth of artificial intelligence microchips to China, authorities said. Chuan Geng, 28, of Pasadena; and Shiwei Yang, 28, of El Monte, were taken into custody on Saturday for their alleged involvement in the illegal overseas export of processing units used in modern computing and artificial intelligence applications, according to a statement from the U.S. attorney's office for the Eastern District of California. Federal prosecutors said both were Chinese nationals, though Geng is a lawful permanent resident of the U.S. Yang, however, was in the country illegally as she had overstayed her visa, according to authorities. Yaoning'Mike' Sun of Chino Hills is charged with acting as an illegal agent of a foreign power and conspiring to advance China-friendly policies in local government. In a criminal complaint, U.S. Justice Department officials alleged the pair had "knowingly and willingly" undercut federal export regulations to conceal illegal shipments to China for nearly three years.

The Biden administration is proposing a new framework for the exporting of the advanced computer chips used to develop artificial intelligence, an attempt to balance national security concerns about the technology with the economic interests of producers and other countries. But the framework proposed Monday also raised concerns of chip industry executives who say the rules would limit access to existing chips used for video games and restrict in 120 countries the chips used for data centers and AI products. Mexico, Portugal, Israel and Switzerland are among the nations that could have limited access. Commerce Secretary Gina Raimondo said on a call with reporters previewing the framework that it's "critical" to preserve America's leadership in AI and the development of AI-related computer chips. The fast-evolving AI technology enables computers to produce novels, make scientific research breakthroughs, automate driving and foster a range of other transformations that could reshape economies and warfare.

The recent emergence of Graphics Processing Units (GPUs) as general-purpose parallel computing devices provides us with new opportunities to develop scalable learning methods for massive data. In this work, we consider the problem of parallelizing two inference methods on GPUs for latent Dirichlet Allocation (LDA) models, collapsed Gibbs sampling (CGS) and collapsed variational Bayesian (CVB). To address limited memory constraints on GPUs, we propose a novel data partitioning scheme that effectively reduces the memory cost. Furthermore, the partitioning scheme balances the computational cost on each multiprocessor and enables us to easily avoid memory access conflicts. We also use data streaming to handle extremely large datasets.

NVIDIA is a technology company that specializes in the design and manufacturing of graphics processing units (GPUs) and system on a chip (SoC) units for the gaming and professional markets. The company was founded in 1993 and is headquartered in Santa Clara, California. NVIDIA's GPUs are widely used in gaming PCs, workstations, and data centers for tasks such as machine learning, data analytics, and scientific simulations. The company also designs and manufactures other products such as Tegra mobile processors, Jetson embedded platform and CUDA parallel computing platform. The company's products are based on its proprietary CUDA architecture, which allows for parallel processing of large amounts of data.

SUNNYVALE, Calif – April 13, 2022 -- Cerebras Systems, the pioneer in high performance artificial intelligence (AI) computing, today released version 1.2 of the Cerebras Software Platform, CSoft, with expanded support for PyTorch and TensorFlow. In addition, customers can now quickly and easily train models with billions of parameters via Cerebras' weight streaming technology. PyTorch is the leading machine learning framework. It is used by developers to accelerate the path from research prototyping to production deployment. As model size increases and as transformer models become more popular, it is essential that machine learning practitioners have access to fast, easy to set up and use compute solutions like the Cerebras CS-2.

Machine Learning and Deep Learning are both becoming well-known phrases in the current era -- but details of the specific tools they require are less ubiquitous. I'd like to discuss one of the most popular Machine Learning tools and how it compares to the others. TensorFlow is probably the most popular Machine Learning tool among researchers today. The Data Incubator calculated that the rating for TensorFlow is nine standard deviations higher than the rating for the second highest machine learning tool, Keras. TensorFlow was written by the Google Brain Team in 2015, and its front end is written in python, while its backend is written in C .

Houston-based ThirdAI, a company building tools to speed up deep learning technology without the need for specialized hardware like graphics processing units, brought in $6 million in seed funding. Neotribe Ventures, Cervin Ventures and Firebolt Ventures co-led the investment, which will be used to hire additional employees and invest in computing resources, Anshumali Shrivastava, Third AI co-founder and CEO, told TechCrunch. Shrivastava, who has a mathematics background, was always interested in artificial intelligence and machine learning, especially rethinking how AI could be developed in a more efficient manner. It was when he was at Rice University that he looked into how to make that work for deep learning. He started ThirdAI in April with some Rice graduate students. ThirdAI's technology is designed to be "a smarter approach to deep learning," using its algorithm and software innovations to make general-purpose central processing units (CPU) faster than graphics processing units for training large neural networks, Shrivastava said.

Artificial intelligence has changed the way we live with innovative technologies. AI has taken a storm in every industry and has a profound impact on every sector of society. The term Artificial intelligence terms were first coined in 1956 at a conference. The advent of the internet helped technology to progress exponentially. Artificial intelligence technology was a stand-alone technology for thirty years, but now the applications are widespread in every sphere of life.

Researchers at the University of Georgia have developed a wearable AI engine that can help visually impaired people navigate the world around them. Users receive audio directions and advisories from a Bluetooth-enabled earphone, while a battery in a fanny pack provides about eight hours of energy. Intel, which provided the processing power for the prototype device, says it's superior to other high-tech visual-assistance programs, which'lack the depth perception necessary to facilitate independent navigation.' Jagadish Mahendran, an AI developer at the University of Georgia's Institute for Artificial Intelligence, was inspired to create the system by a visually impaired friend. 'I was struck by the irony that, while I have been teaching robots to see, there are many people who cannot see and need help,' he said. Using OpenCV's Artificial Intelligence Kit, he developed a program that runs on a laptop small enough to stow in a backpack, linked to Luxonis OAK-D spatial AI cameras in a vest jacket that provide obstacle and depth information.

Global demand for machine learning hardware is dramatically outpacing current computing power supplies. State-of-the-art electronic hardware, such as graphics processing units and tensor processing unit accelerators, help mitigate this, but are intrinsically challenged by serial data processing that requires iterative data processing and encounters delays from wiring and circuit constraints. Optical alternatives to electronic hardware could help speed up machine learning processes by simplifying the way information is processed in a non-iterative way. However, photonic-based machine learning is typically limited by the number of components that can be placed on photonic integrated circuits, limiting the interconnectivity, while free-space spatial-light-modulators are restricted to slow programming speeds. To achieve a breakthrough in this optical machine learning system, the researchers replaced spatial light modulators with digital mirror-based technology, thus developing a system over 100 times faster.