Open.Intel’s Post

Researchers at Peking University recently developed a new reconfigurable neuromorphic computing platform that integrates sensing and computing functions in a single device. #photonicmemristors #neuralnetwork #computervision

Quantum Computing by Mathematics: achieving tomorrow's technology with today's tools. In continuation of the previous posts, despite the advances to achieve the power of quantum computing at temperatures such as 1 Kelvin and efforts to use photonic and laser styles to achieve this power at room temperature, it is still far from reaching the level of using quantum computers in consumers' homes with reasonable costs. As mentioned, computing styles based on artificial intelligence and neural, cognitive, and neuromorphic computing are also being developed. Of course, these methods seek to imitate one hundred percent the behavior of human brain calculations in machines, so that they can convey most of the possible abilities and states. Meanwhile, using computational tools from various branches of mathematics such as algebra, analysis, and geometry is a completely different approach. Changing the basis of calculations from binary scalar states to vectors, matrices and other high-dimensional tools such as tensors and manifolds is a style that is very close to quantum states and gates. Of course, the origin of this style was trying to simulate states and quantum computing with mathematical algorithms on classical computers, as well as managing big data and multidimensional data with high dimensions. As a person who has studied, investigated, and taught various branches of mathematics and statistics for many years, and interested and active in computer science for a long time, the use of mathematical and statistical tools is a cost-effective way to achieve the power of quantum computing on the platform of today's classical computers. The simultaneous use of mathematical computing intelligence styles and the maximum power of advanced technologies of hardware companies such as Nvidia, AMD, and Intel based on parallel and distributed processing can promise to achieve high processing power similar to quantum computers on today's tools. #quantum #quantumcomputing #tensor #manifold #mathematics #nvidia #amd #intel

Today Intel introduced the world’s largest neuromorphic system surpassing the billion neurons mark. Code-named Hala Point, the 1.15 billion neuron system was deployed at Sandia National Laboratories for advanced brain-scale computing research. The organization will focus on solving scientific computing problems in device physics, computer architecture, computer science, and informatics.   Hala Point improves on Intel’s first-generation large-scale research system, Pohoiki Springs, with over 10x more neuron capacity and up to 12x higher performance on AI workloads. It can support up to 20 quadrillion operations per second, or 20 petaops, with an efficiency exceeding 15 trillion 8-bit operations per second per watt (TOPS/W) when executing conventional deep neural networks. This rivals and exceeds levels achieved by GPU and CPU architectures.   Currently, Hala Point is a research prototype that will be used to advance the capabilities of future commercial systems. This could lead to breakthroughs in future real-time continuous learning for AI applications such as scientific and engineering problem-solving, logistics, smart city infrastructure management, LLMs, and AI agents.   Congratulations to the Neuromorphic Computing Lab at Intel Labs for this leap forward in sustainable AI deployment as well as advancing the state-of-the-art to enable a new level of scientific research in neuromorphic computing.   Learn more: https://lnkd.in/dYsGp_Zr   #iamintel #Neuromorphic #ArtificialIntelligence #LLM #GenerativeAI

This article introduces a new optical computing technique called "diffraction casting," which aims to overcome the limitations of traditional optical computing by leveraging the properties of light waves. This approach is more spatially efficient and flexible compared to older methods like shadow casting. Researchers used this method to perform logical operations on small images, demonstrating its potential for image processing and other data representations. While still in early development, diffraction casting could eventually complement existing computing technologies, with possible applications in machine learning and quantum computing. Please continue reading the full article under the following link: https://lnkd.in/ewp9iy5U #materials #materialsscience #materialsengineering #computationalchemistry #modelling #chemistry #researchanddevelopment #research #MaterialsSquare #ComputationalChemistry #Tutorial #DFT #simulationsoftware #simulation

Intel’s Aurora achieves exascale to become the fastest AI system: Intel, in partnership with Argonne National Laboratory and Hewlett Packard Enterprise (HPE), has announced that its Aurora supercomputer has surpassed the exascale computing threshold, achieving speeds of 1.012 exaflops and becoming the fastest AI-focused system. The supercomputer, designed as an AI-centric system, has demonstrated groundbreaking work in various scientific fields, including neuroscience, particle physics, and drug discovery. It is powered by the Intel Data Center GPU Max Series and supported by a suite of software tools to enhance developer flexibility and system scalability. Intel's commitment to advancing HPC and AI is evident in its expansion of the Tiber Developer Cloud and the deployment of new supercomputers integrated with Intel technologies, which are expected to transform scientific research in fields such as climate change modeling and fusion energy.

Neuromorphic computing: Neuromorphic computing is a method of computer engineering in which elements of a computer are modeled after systems in the human brain and nervous system. The term refers to the design of both hardware and software computing elements. Neuromorphic computing is sometimes referred to as neuromorphic engineering. Neuromorphic engineers draw from several disciplines -- including computer science, biology, mathematics, electronic engineering and physics -- to create bio-inspired computer systems and hardware. Of the brain's biological structures, neuromorphic architectures are most often modelled after neurons and synapses. This is because neuroscientists consider neurons the fundamental units of the brain. https://lnkd.in/gXNXb77m

A decade ago, we could not have envisaged the reality of cognitive computing, but recent advances in physics and nanotechnologies have made it possible. In this article you'll find some examples of the areas where computing paradigms are rapidly changing. #DigitalTransformation #technology #innovation #supercomputing #quantumcomputing

I wrote this earlier this year. It is interesting to watch progress, particularly on the quantum tech front with Google’s Willow breakthrough, and Quantum Computing Inc’s operational photonic quantum computing. Quantum, like artificial intelligence will be transformational, especially when they are combined! #Quantumcomputing #techtrends #computers Google Quantum Computing, Inc. #futuretrends Computational Capabilities That Will Transform the World By Chuck Brooks https://lnkd.in/eSeNtCkS

🦸The #Aurora #Supercomputer has broken records by achieving an astounding 1.012 exaflops of computational power, making it the world’s fastest AI system dedicated to open science. 📌This breakthrough, made possible by #Intel, #Argonne National Laboratory, and Hewlett Packard Enterprise, is set to revolutionize scientific discovery across various domains, including climate science, particle physics, and drug discovery, by leveraging the power of generative AI models. 🐞With 10,624 compute blades and 84,992 HPE Slingshot Fabric Endpoints, Aurora is a true marvel of state-of-the-art hardware components. The system's scalability and efficiency are impressive, as demonstrated by its second-place finish in the HPL Benchmark and third-place finish in the HPCG Benchmark, delivering an impressive 5,612 TF/s. Aurora is a game-changer, and we can't wait to see the scientific advancements it will bring to the world. #AuroraSupercomputer #AI #OpenScience #HPC

Benchmarking quantum machine learning for surface crack detection reveals significant potential for improved accuracy and efficiency on current quantum computing