Skip to main content
Blog

AI/ML Innovations at RISC-V Summit North America: A Track to Watch

By October 8, 2024No Comments

Registration prices for RISC-V Summit North America 2024 increase after Oct 11. Register today to sit in on AI/ML presentations and more.

Artificial Intelligence (AI) and Machine Learning (ML) are revolutionizing industries worldwide, and their impact is only growing. These technologies, driving everything from automation to data analysis, are integral to the future of computing. The increasing popularity of AI/ML has put significant pressure on the underlying hardware, leading to a surge in demand for specialized architectures that can accelerate AI workloads efficiently.

At the heart of this evolution is RISC-V, an open standard instruction set architecture (ISA) that’s changing the landscape of AI development. AI and ML innovations will be front and center at the RISC-V Summit North America 2024. With presentations from leading industry experts, the summit will showcase how RISC-V is being leveraged to build advanced AI accelerators, optimize generative AI models, and create toolchains for AI workloads.

As AI/ML applications expand, the need for cost-effective, high-performance compute systems becomes more critical. Attendees at this year’s RISC-V Summit will hear firsthand how the RISC-V ecosystem is addressing these challenges, offering a unique opportunity to stay ahead of the curve in AI development.

Here are some key AI/ML presentations you won’t want to miss:

Featured AI/ML Presentations:

  • Say Goodbye to Fear, Uncertainty, and Doubt: Innovate with Codasip Studio Fusion

Presented by: Keith Graham, Codasip

Codasip Studio Fusion helps AI developers eliminate the fear and uncertainty associated with custom compute by providing a powerful development platform. This session will explore how Codasip’s technology enables fast, confident innovation in AI.

  • The Benefits of Building New AI Accelerators with RISC-V

Presented by: Cliff Young & Martin Maas, Google DeepMind

This talk dives into Google’s experience building AI accelerators and how RISC-V is transforming the accelerator development process by lowering barriers to entry and reducing risk.

  • Lessons Learned in Using RISC-V for Generative AI and Where We Can Go from Here

Presented by: Jin Kim, Esperanto Technologies

As generative AI models grow exponentially in size, overcoming the “memory wall” and “compute wall” becomes a pressing challenge. Learn how Esperanto Technologies is leveraging RISC-V to extend traditional AI hardware to meet the demands of generative AI.

  • Building Tool Chains for RISC-V AI Accelerators

Presented by: Jeremy Bennett, Embecosm

This session will share insights from Embecosm’s work building toolchains for massively parallel RISC-V AI chips. Attendees will learn about the challenges faced and solutions found in running AI workloads on RISC-V, using popular frameworks like PyTorch and TensorFlow.

  • LLM Inference on RISC-V Embedded CPUs

Presented by: Yueh-Feng Lee, Andes Technology

Explore how RISC-V CPUs are optimizing large language model (LLM) inference for resource-constrained environments. This session demonstrates the potential of deploying AI models efficiently on edge devices.

  • Bridging the Gap: Compiling and Optimizing Triton Kernels Onto RISC-V Targets Based on MLIR

Presented by: Aries Wu, Terapines Technology Co., Ltd.

This talk presents a comprehensive AI software stack solution for RISC-V, using Triton kernels and MLIR/LLVM-based AI compiler infrastructure to optimize AI workloads for non-GPU architectures.

The Time to Register is Now!

The RISC-V Summit North America is a must-attend event for anyone looking to stay at the forefront of AI/ML innovation. With ticket prices increasing on October 11, now is the perfect time to secure your spot and gain insights into how RISC-V is shaping the future of AI.

Don’t miss out on these groundbreaking presentations and more. Register today and join the community that’s driving the future of computing! Register here.