Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

YouTube

WarpDrive: Orders of Magnitude Faster Multi-Agent Deep Reinforcement Learning on a GPU

MLOps World: Machine Learning in Production via YouTube

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore the cutting-edge WarpDrive framework for multi-agent deep reinforcement learning in this 56-minute conference talk from MLOps World: Machine Learning in Production. Dive into the world of GPU-accelerated RL with Stephan Zheng, Lead Research Scientist at Salesforce, and Tian Lan, Senior Applied Scientist at Salesforce. Learn how WarpDrive enables orders of magnitude faster reinforcement learning by running end-to-end on a GPU. Discover the framework's latest features, including automatic GPU utilization tuning, distributed training on multiple GPUs, and sharing multiple GPU blocks across simulations. Understand how these advancements allow for linear throughput scaling with the number of devices, supporting up to millions of agents. Gain insights into the utility functions that improve user experience and facilitate rapid implementation of RL workflows. Explore real-world applications of reinforcement learning, from superhuman gameplay to optimizing data center cooling, nuclear fusion control, and economic policy analysis.

Syllabus

WarpDrive: Orders of Magnitude Faster Multi-Agent Deep RL on a GPU

Taught by

MLOps World: Machine Learning in Production

Reviews

Start your review of WarpDrive: Orders of Magnitude Faster Multi-Agent Deep Reinforcement Learning on a GPU

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.