➀ The Technical University of Graz (TU Graz) is developing a real-time lightning risk assessment system to improve safety at outdoor events and construction sites. ➁ The system uses a network of field meters and combines data from lightning location systems and weather radar to predict lightning strikes. ➂ The research aims to reduce downtime and enhance safety by providing more accurate forecasts.
Recent #machine learning news in the semiconductor industry
➀ The current state-of-the-art machine learning applications use deep neural network models that have become so large and complex that they have exceeded the limits of traditional electronic computing hardware.
➁ Researchers at MIT and other institutions have developed a new type of photonic chip that can overcome these barriers.
➂ The optical device can complete the key calculations of machine learning classification tasks in less than half a nanosecond while achieving an accuracy rate of over 92%.
➀ Jensen Huang discusses the current state of AI development and the issue of AI hallucinations; ➁ He emphasizes the rapid increase in AI computing power demand over the past decade; ➃ Huang shares personal anecdotes about his career and his early promises.
➀ The Fraunhofer Institute for Production Systems and Design Technology IPK presents research projects and solutions addressing the challenges of modern manufacturing; ➁ The focus is on data management, flexible manufacturing processes, machine efficiency, human resource support, and sustainable production; ➂ The article highlights the importance of holistic system solutions and interdisciplinary collaboration in research and innovation.
➀ A learning robot has been developed at the TU Wien; ➁ The robot can imitate a cleaning process and apply it to different situations; ➂ The robot learns through demonstration and can clean various surfaces despite being trained on a single edge.
➀ Alex Bronstein, known for groundbreaking innovations, starts his professorship at the Institute of Science and Technology Austria (ISTA) with a focus on Machine Learning in biowissenschaften. ➁ Bronstein has a background in both academic research at the Technion, Israel, and industry at Intel, where he developed 3D sensing technology. ➂ He aims to expand the boundaries of machine learning for applications in biowissenschaften and contribute to strategic directions in structural and cell biology, as well as single-cell analyses.
➀ Microchip introduces the PIC64HX microprocessor family for advanced intelligent edge designs; ➁ The family offers high-performance AI and ML capabilities, TSN Ethernet connectivity, and post-quantum security; ➂ The PIC64HX is designed to power the future of intelligent edge applications.
➀ Cloud Architect with an average annual salary of $174,984; ➁ Network Security Engineer with an average annual salary of $160,471; ➂ Data Scientist, Machine Learning Engineer, Cybersecurity Engineer, Software Engineer, Penetration Tester, AI Research Scientist, DevOps, and Information Security Analyst are among the highest paying jobs in tech.
➀ An AI researcher uses Google AI Studio to convert a 35-second screen recording into a spreadsheet; ➁ The process costs less than a cent due to the low token rate of Google AI Studio; ➂ The technology could save thousands of hours of manual labor.
➀ Researchers from the Leibniz Institute for Astrophysics Potsdam (AIP) and the Institute of Cosmos Sciences at the University of Barcelona (ICCUB) have used a novel machine learning model to process observation data from 217 million stars of the Gaia mission efficiently. The results are comparable to conventional methods for determining stellar parameters. The new approach opens up exciting possibilities for mapping properties like interstellar extinction and metallicity across the Milky Way, contributing to understanding the stellar populations and the structure of our galaxy. ➁ The third data release of the Gaia satellite by the European Space Agency ESA provided access to improved measurements for 1.8 billion stars, a vast amount of data for studying the Milky Way. Efficient analysis of such a large dataset, however, presents a challenge. The study published now investigates the use of machine learning to determine important stellar properties based on Gaia's spectrophotometric data. The model was trained on high-quality data from 8 million stars and achieved reliable predictions with low uncertainties. ➂ The machine learning technique, 'Extreme Gradient-Boosted Trees,' enables the determination of precise stellar properties like temperature, chemical composition, and interstellar dust extinction with unprecedented efficiency. The developed machine learning model, SHBoost, completes its tasks, including model training and prediction, within four hours on a single graphics processor - a process that previously required two weeks and 3000 high-performance processors.
➀ The Raspberry Pi AI Camera offers basic AI capabilities for $70; ➁ It performs well in object detection and pose estimation; ➂ It can be used as a standard Raspberry Pi camera; ➃ The camera has a lower computational power compared to the Raspberry Pi AI Kit.
➀ The article presents a novel technique for the on-demand nanoengineering of in-plane ferroelectric topologies; ➁ The technique involves a combination of scanning probe microscopy and machine learning algorithms; ➂ The research demonstrates the ability to create complex patterns and functionalities by manipulating the polarization in ferroelectric materials; ➃ The findings could lead to new applications in electronics, sensors, and energy storage.
➀ Nuvoton Technology launches the NuEzAIM55M1 development board; ➁ The board is designed to simplify endpoint AI development; ➂ It is built on the Arm CortexM55 core and includes a model training tool.
➀ The cost of training AI models has surged in the past year, as indicated by data from Epoch AI; ➁ The training cost for Gemini, a large language model from Google, ranges from $30 to $191 million; ➂ OpenAI's ChatGPT-4, released in 2023, had a technical creation cost between $41 million and $78 million.
➀ Sentienz's Akiro IoT platform addresses challenges in e-transportation with advanced analytics and MQTT+; ➁ Akiro optimises IoT outcomes through messaging, analytics, and AI/ML; ➂ The platform supports smart meters, EV chargers, and battery monitoring systems, enhancing EV charging infrastructure and data management.
➀ Cadence's Tensilica HiFi 5 DSPs are integrated into NXP's new automotive audio DSP family; ➁ The integration supports advanced audio capabilities for next-generation software-defined vehicles; ➂ This development caters to the increasing demand for sophisticated audio processing.
➀ Google has rebranded TensorFlow Lite to LiteRT, allowing for ML and AI model deployment on Android, iOS, and embedded devices; ➁ LiteRT supports multiple frameworks including PyTorch, JAX, and Keras; ➂ The name change is part of the Google AI Edge suite and will be progressively rolled out with the LiteRT name appearing more in developer documentation.
➀ Meta Platforms identifies a widening gap between computing power and interconnect bandwidth at the 2022 OCP Global Summit. ➁ Broadcom proposes SCIP (Silicon Photonics in Package) as a solution to bridge this gap, focusing on low-cost, high-performance, and low-power interconnects. ➂ SCIP utilizes TSV technology and detachable optical connectors to achieve shorter interconnect distances and higher energy efficiency, targeting AI and machine learning applications.