Patent Leaderboard
USPTO Patent Rankings Data through Dec 31, 2025
Home › Methods of operating a graphics processing unit (GPU) to train a deep neural network using a GPU local memory and related articles of manufacture

Methods of operating a graphics processing unit (GPU) to train a deep neural network using a GPU local memory and related articles of manufacture

US Patent 11599798 · Granted Mar 7, 2023

Assignee

Inventors

View full patent text on Google Patents →

More Patents by These Inventors

Compute optimizations for neural networks2025
Multicore processor with each core having independent floating point datapath an2025
Method of manufacturing spliced display screen and spliced display screen2025
Roller device2025
Coordinated control method and system for active support of new energy power sta2025
Compute optimizations for low precision machine learning operations2025
Training with adaptive runtime and precision profiling2025
Flexible double-sided display screen and manufacturing method thereof2025
Dynamic precision for neural network compute operations2025
Ionospheric disturbance information generation methods and systems2025
Display panel and method for preparing display panel2025
Array substrate including peeling prevention and anti-reflective functional laye2025