Mark de Burgh

Learn how the ANSVIS AI platform can be integrated into existing LabVIEW and manufacturing systems to conquer challenging computer vision problems. Tuan Nghia Nguyen, PhD FIEAust CPEng NER guides us through the process of image data collection and labelling, configuring and training the ANSVIS AI model, deployment, and third-party integration. https://lnkd.in/gpRx6tdY #labview #gdevcon #gdevconanz #labviewfriends #labviewcon #labviewconference #labviewdevelopers #techcon #teststand

10

In this episode, host Paul Spain dives into the latest processors from Intel and HP. Paul examines the impact of local AI processing and the evolving landscape of chip architectures. Learn about the implications for performance, battery life, and software compatibility as the tech industry races towards a future of enhanced computing power. #localai #cloudbased

4

Argonne National Laboratory researchers have developed a new workflow that uses #HPC and #AI at the edge to enable real-time streaming ptychographic imaging at the Advanced Photon Source. The team used ALCF supercomputing resources for online neural network training and an embedded GPU device at the APS for real-time imaging analysis. The new workflow provides a scalable solution to the ever-growing data rate with state-of-the-art ptychography, and can be easily extended to work with other imaging techniques at light sources and advanced electron microscopes. Read the story: https://lnkd.in/gS2GwpSX To learn more, see the team's paper, "Deep learning at the edge enables real-time streaming ptychographic imaging," in Nature Communications: https://lnkd.in/dvmrS-qG

11

My Academic training was at Monash University, Melbourne Australia 🇦🇺 took Engineering and Computing with 1st Class Honours, thru Panasonic Scholarship, MEI Matsushita Electric Industrial based in Kadoma, Osaka Japan 🇯🇵 In those days, with this qualification, we can join Engineers Australia and Computer Society of Australia as graduate member, Professional Career Development and Network. What I see the handicap of almost all Singapore Universities Computing graduates, none "fuse" hard Engineering with Computer Science. This will spell challenges for these fresh graduates, especially from Computer Science, since ChatGPT can do Coding, why need another CompSci new hire? For pure Engineering, their jobs are perfectly secured in Infrastructure related sectors, PSA, Singapore Airline Engineering, Changi Airport, ST Engineering, Semiconductor Fabs, Oil and Gas, Jurong Island, Power Generation Plants, Building and Construction 🚧 What I see in extreme demand is Hardware Software Artificial Intelligence = Machine Intelligence or the Fusion of AI as embodiment into Physical Digital Devices from Robotics to Automation and perhaps Autonomous Self-driving cars like Waymo. The new SkillsFuture of this cohort has to be able to put AI into Factories, Restaurants, Hospitals, almost everywhere, more than just into workflows, but Machines that have Intelligence. Come to think of it, you believe coding as Digital form as control the movement of Electric Motors? In other words, Digital-to-Analogue or DAC, just by plugging in RJ45 into the voltage terminals? Think again ...... I see many fresh graduates will struggle to secure jobs at Amazon, for example, think of it as the most Advanced Automated Warehouse with everything AI controlling those conveyor belts, storage systems, hoist-n-lift mechanism, with little robots running around collecting stuff we buy !!!!! Or for that matter, just look at Nvidia, how the World's most powerful AI Chip called Blackwell with half-size of a Credit Card that has over 208 billion transistors. New AI Engineers will work with Chips 🍟 By the way, what's 208 billion or 208,000,000,000? Sony's World's 1st transistor Radio launched in 1955 was called TR-55 with a size of an 8-inch tablet and had 5 transistors on-board, licensed from Bell Labs - Talk about listening thru 95.8 📻 with Jay Chou on call. I think this massive CompScience cohort will struggle as they lack Engineering integration, most importantly ill-equipped with Hardware capabilities and competenies to engage AI Chips and Digital Machines. Therefore, I hope senior students will read this and start your internship early, add Hardware hands-on experience, beyond only knowing the keyboard, as ChatGPT will Disrupt your FutureReady, even before you start your short-lived career 😉

30

7 Comments

This paper introduces a novel #transformer-#based #deep #recurrent #architecture (#TDRA) for #automatic #modulation #classification (#AMC), tailored for #intelligent #reflecting #surface (#IRS)-assisted #Internet #of #Things (#IoT) networks, which significantly improves IoT device performance in NLoS scenarios. In TDRA, the existing recurrent models, #long-#short-#term #memory (#LSTM), and #gated-#recurrent-#unit (#GRU) are suitably revamped with a transformer-based approach and termed as transformer-based #LSTM (#T-#LSTM) and #transformer-#based #GRU (#T-#GRU). ----Debbarni Sarkar, @Yogita, Satyendra Singh Yadav (He/Him), @Linga Reddy Cenkeramaddi, Om Jee Pandey More details can be found at this link: https://lnkd.in/ghZevX6h

11

2 Comments