Presentation
State of the Art GPU Code Generation with Reinforcement Learning
DescriptionWriting high performance parallel code takes a long time and offers a steep learning curve. Today's LLMs are helpful, but not quite up to the task. We create an agent architecture with a fine tuned model that achieves state of the art results, allowing anyone to write code for GPUs effectively.
Presenter
Event Type
Community Engagement and Support
TimeTuesday, 18 November 20254:30pm - 5:00pm CST
LocationBooth 3537 - SCinet Theater