Patent Leaderboard
USPTO Patent Rankings Data through Dec 31, 2025
Home › Learning a truncation rank of singular value decomposed matrices representing weight tensors in neural networks

Learning a truncation rank of singular value decomposed matrices representing weight tensors in neural networks

US Patent 11238346 · Granted Feb 1, 2022

Estimated economic value: $28,075,000 ⓘ

Assignee

Inventors

View full patent text on Google Patents →

More Patents by These Inventors

Reference signal design based on semi-uniform pilot spacing2024
Quantizing neural networks with batch normalization2024
Path detection for autonomous machines using deep neural networks2023
Determining layer ranks for compression of deep networks2023
Generating a sparse feature vector for classification2022
Feature selection for retraining classifiers2022
Incremental learning for dynamic feature database management in an object recogn2022
Reference signal design based on semi-uniform pilot spacing2021
Filter specificity as training criterion for neural networks2019
Autonomously performing default operations based on current operating contexts2019
Knowledge-graph biased classification for data2019
Context-based priors for object detection in images2019