Patent №
US 11,625,554
Granted
2023-04-11
Filed 2019
Owner
INTERNATIONAL BUSINESS MACHINES CORPORATION
Lab
AI components
5
ml · vision · speech · evo · hardware
Assignment
Recorded
Dataset
AIPD
2023_r1 edition
Application
16266755
A training method, system, and computer program product include training a neural network including using two-sided ReLU as a non-linear function or norm-pooling as a non-linear function and increasing a confidence gap.
Machine learningVisionSpeechEvolutionary computationAI hardwareG06F 17/11G06F 17/16G06F 18/2148G06N 3/0464G06N 3/048G06N 3/08G06N 3/088G06N 3/09+8 more
AI classification
Machine learning1.00
Vision1.00
AI hardware0.99
Speech0.95
Evolutionary computation0.91
Planning0.15
Natural language0.10
Knowledge representation0.00
Ownership
INTERNATIONAL BUSINESS MACHINES CORPORATION
assignment · 482450429
Assignors
QIAN, HAIFENG, WEGMAN, MARK
On an employer assignment, the assignors are typically the inventors.
From the same owner
METRIC LEARNING PREDICTION OF SIMULATION PARAMETERSUS 12,691,869 · 2026DETERMINATION AND ALLEVIATION OF ROOT CAUSES FOR EMISSION OF POLLUTANTS FROM VEHICLESUS 12,693,686 · 2026TRAINING AND USING A SENTIMENT MACHINE LEARNING MODULE TO RECEIVE AS INPUT HAPTIC METRIC VALUES TO DETERMINE A SENTIMENT SCORE FOR TEXT TO PROVIDE TO AN INTERACTIVE PROGRAMUS 12,693,779 · 2026DOMAIN-SPECIFICITY PREDICTION FOR NATURAL LANGUAGE PROCESSINGUS 12,694,208 · 2026USER INTERFACE AUTOMATION USING NATURAL LANGUAGEUS 12,694,222 · 2026NATURAL LANGUAGE DATA GENERATION USING AUTOMATED KNOWLEDGE DISTILLATION TECHNIQUESUS 12,694,236 · 2026NEURAL NETWORK SYSTEMS FOR ABSTRACT REASONINGUS 12,694,281 · 2026ACCELERATING INFERENCE OF NEURAL NETWORK MODELS VIA DYNAMIC EARLY EXITSUS 12,694,283 · 2026