Michael Wong
Codeplay Technology VP
Michael is the Chair of the C++ Embedded Development Committee SG14 and the Machine Learning Committee SG19, the Chair of the C++ Language Direction Evolution Committee, the Vice President of Research and Development at Codeplay, and the Head of the Canadian Delegation to the C++ Standards Board. Michael's extensive experience in the areas of C++ Parallel Computing, High Performance Computing, and Machine Learning has led the development of the C++ Heterogeneous Programming Language (SYCL) Standard for the Development of C ++ Applications on GPUs and GPUs. He led the development of the C++ Heterogeneous Programming Language (SYCL) standard for GPU application development. He has deep research and insights into the underlying performance optimization of Tensorflow. His specific work covers parallel programming, neural networks, computer vision, autonomous driving, etc. Michael was a senior technical expert at IBM, leading the development of IBM XL C++ compiler and XL C compiler.
Topic
Unlocking AI's Potential: Navigating the Evolving Landscape of AI Hardware and Software
Artificial intelligence (AI) is undergoing a period of rapid innovation, driven by advances in both hardware and software. On the hardware front, we are seeing the development of new AI accelerators such as GPUs, TPUs, FPGAs, and ASICs. These accelerators offer significant performance improvements over traditional CPUs, but they also pose challenges for compatibility and optimization. On the software front, we are seeing the development of new AI frameworks such as PyTorch, TensorFlow, JAX, Pallas, and Triton. These frameworks provide developers with the tools they need to build and deploy AI models, but they also need to be continuously updated to support new hardware capabilities. In this talk, we will discuss the challenges and opportunities of the evolving AI hardware and software landscape. We will examine the different types of AI accelerators and frameworks that are available, and we will discuss the importance of compatibility and optimization. We will also discuss the role of open-source initiatives in promoting innovation and collaboration in the AI ecosystem. This talk is intended for anyone who is interested in learning more about the AI hardware and software landscape. Whether you are a developer, researcher, or decision-maker, this talk will provide you with the insights you need to navigate this complex and rapidly evolving field.
