The top conference AAAI 2024 has successfully accepted a recent paper from the AMD. This blog shares about keys of an efficient depth pruning method for both CNN and Vision Transformer, and how this method helps outstanding compression performance across various AI models. 

AMD, through its powerful GPU computing platforms like AMD Instinct™ accelerators, AMD Radeon™ GPUs, and AMD ROCm™ software, offers developers and the AI community alternative solutions for AI training and inferencing. Meanwhile, with the establishment of an open AI software ecosystem, AMD allows more partners to participate in contributing variable AI acceleration solutions, particularly for large AI models on AMD GPU platforms.

AMD is excited to kick off the Pervasive AI Developer Contest; a competition designed to spark innovation amongst AI developers.  Showcase your creativity and technical expertise on chosen AMD platforms and participate in one of three categories: Generative AI, Robotics AI, and PC AI.

I’m excited to share that today, AMD has completed the acquisition of Nod.ai - further expanding our open-source AI software capabilities. We are thrilled to welcome Nod.ai’s talented team of AI software experts to the AMD Artificial Intelligence Group (AIG).   

We live in an age of rapid technological advancement. Machine learning has become a term that has permeated every facet of our lives. From the personalized recommendations suggested to you when you sit down to stream a movie, to self-driving cars navigating our streets, machine learning is what makes these innovations possible. But what exactly is machine learning, and how is it transforming the world around us?

AMD knows a lot about supercomputing. In fact, AMD powered supercomputers hold 2 of the top 5 spots on the Top500 list – including the first system to bring the world into the era of exascale (a computer capable of performing a quintillion or 10^18 floating point operations per second), holding an HPL (High Performance LINPACK) score that is nearly triple that of ­­­the second-place listing. Though, fewer know how much work it is to get to the point of flexing a system’s computing muscle to a worldwide stage. There is a ton of work that happens behind the scenes before any of these massive systems can “open its doors” to production workloads. One of the main milestones for any supercomputer is called acceptance. Let’s break down what leads a supercomputer to be accepted and find out what that term really means. Then, learn how AMD contributes to building the most performant systems in the world and some of the science they enable.

Read how Moreh and KT were able to supercharge their AI workloads with AMD Instinct™ MI250 accelerators.

AMD Composable Kernel library: create highly efficient fused kernels for AI applications with just a few lines of code.

Learn how AMD is addressing the challenge of delivering high performance backend kernels for a wide range of fused tensor operators at rapid pace with the AMD Composable Kernel (CK) library. Chao Liu, Jing Zhang [1]

Today, November 8, 2021, marks the beginning of a new era in GPU-accelerated computing with the announcement of the latest AMD Instinct™ MI200 series accelerators, featuring the world’s fastest – the Instinct MI250X1. The GPU technology being used to power the Frontier supercomputer, which is expected to be the first US exascale system —at 1.5 exaFLOPs— will be available for leading HPC and AI practitioners to use for new industry and research discoveries.