Fueling the AI Revolution with Gaming
summary
NVIDIA explains how GPUs and software actually accelerate AI: CUDA as the programming heart, DGX servers with multiple cards and containerized stacks on NGC, and edge modules like Jetson TX2 for 1–1.5 trillion ops in a credit‑card board. Claims include Volta tensor cores multiplying throughput, Pegasus at ~320 TOPS for autonomous perception, 500+ GPU‑ready apps, and a robotics platform Isaac (VR Holodeck + reinforcement learning) used for one‑shot imitation and parkour. Resources: free frameworks, DLI courses and 200+ hands‑on classes.
AI-written summary · about
chapters (12)
- 0:00FUELING THE AI REVOLUTION WITH GAMING
- 6:24Definitions
- 7:47HOW GPU ACCELERATION WORKS Application Code
- 17:22REINFORCEMENT LEARNING & ROBOTS
- 27:07NVIDIA DEEP LEARNING SOFTWARE PLATFORM
- 28:41NVIDIA DEEP LEARNING SDK UPDATE
- 30:50SINGLE UNIVERSAL GPU FOR ALL ACCELERATED WORKLOADS
- 35:13THREE STEPS TO DEEP LEARNING WITH NGC
- 35:54GPU DEEP LEARNING IS A NEW COMPUTING MODEL
- 36:36JETSON TX2
- 38:16NVIDIA Deep Learning Institute
- 39:41COME DO YOUR LIFE'S WORK JOIN NVIDIA
related talks
28mDeveloper Essentials for the New Era of AI
20mAI : Solving challenges and driving Innovation across Industries
16mUsing AI and Computer Technologies to Enhance IT Infrastructure
23mFrom Gaming to the Metaverse
20mAccelerating AI pipelines w/ NVIDIA on AWS: An autonomous driving case
21mAnalyst Notebook
24mKeynote: Next-generation frameworks for Large-scale AI
30mAI is fuelling all areas of business