Microchip Technology has made a significant move in the AI landscape with the acquisition of Neuronix AI Labs. The strategic decision aims to enhance Microchip’s capabilities in developing power-efficient, AI-enabled edge systems utilizing field programmable gate arrays (FPGAs). Neuronix AI Labs specializes in neural network sparsity optimization technology, which significantly reduces power consumption, size, and computational requirements for various tasks like image classification and object detection while maintaining high accuracy.
This acquisition comes at a time when Mid-range PolarFire® FPGAs and SoCs from Microchip are already renowned for their low power consumption, reliability, and security features. By integrating Neuronix AI Labs’ optimization technology, Microchip aims to enable cost-effective, large-scale edge deployments of components tailored for computer vision applications that face constraints related to size, power, and cost. This move is expected to result in a substantial increase in AI/ML processing power on low and mid-range FPGAs.
Bruce Weyer, Corporate Vice President of Microchip’s FPGA business unit, highlighted the significance of this acquisition in enhancing power efficiency for FPGAs and SoCs deployed in intelligent edge systems utilizing AI/ML algorithms. The combination of Neuronix technology and Microchip’s VectorBlox™ design flow is anticipated to improve neural network performance efficiency and deliver outstanding performance in terms of GOPS/watt on PolarFire FPGAs and SoCs.
Moreover, the neural network sparsity optimization technology offers a user-friendly approach for non-FPGA designers to leverage powerful parallel processing capabilities using standard AI frameworks without extensive knowledge of FPGA design flow. By integrating Neuronix AI intellectual property with Microchip’s compilers and software design kits, AI/ML algorithms can now be implemented on customizable FPGA logic without the need for intricate FPGA fabric expertise. This approach also allows for seamless updates and upgrades of Convolutional Neural Networks (CNNs) without requiring hardware reprogramming.
Yaron Raz, CEO of Neuronix AI Labs, expressed excitement about the opportunity to scale and align their advanced neural network acceleration architectures and algorithms with Microchip’s industry-leading FPGA portfolio, known for setting benchmarks in power efficiency. The collaboration between the two entities is expected to bring transformative advancements in size, power, performance, and cost-effectiveness within the AI landscape.
Overall, the acquisition of Neuronix AI Labs by Microchip Technology marks a significant milestone in advancing the capabilities of edge systems powered by FPGAs. The integration of neural network sparsity optimization technology is poised to revolutionize the landscape of AI-enabled applications by addressing key challenges related to power efficiency, size constraints, and computational requirements.
