Inventor

LE QUOC V

US109 patents

⚠️ This page may combine multiple inventors who share the name “LE QUOC V”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

LE QUOC V

GOOGLE LLC

Training a document classification neural network

Generating representations of input sequences using neural networks

Generating author vectors

Systems and methods for progressive learning for machine-learned models to optimize training speed

Processing text sequences using neural networks

Generating parameter values for recurrent neural networks

Neural architecture search

Neural network programmer

Speech recognition with attention-based recurrent neural networks

Generating target sequences from input sequences using partial conditioning

Training machine learning models using teacher annealing

Training neural networks using data augmentation policies

Training a document classification neural network

Computationally efficient neural network architecture search

Compound model scaling for neural networks

Generating vector representations of documents

Implicit bridging of machine learning tasks

Generating author vectors

End-to-end text-to-speech conversion

Neural machine translation systems with rare word processing

Neural machine translation systems with rare word processing

Systems and methods for progressive learning for machine-learned models to optimize training speed

Generating target sequences from input sequences using partial conditioning

Speech recognition with attention-based recurrent neural networks

Recurrent neural networks with rectified linear units

Generating target sequences from input sequences using partial conditioning

Neural architecture search for convolutional neural networks

Generating vector representations of documents

Contrastive pre-training for language tasks

Compound model scaling for neural networks

Training a document classification neural network

Systems and methods for machine-learned models having convolution and attention

Implicit bridging of machine learning tasks

Neural symbolic reader

Training neural networks using priority queues

Device placement optimization with reinforcement learning

Granular neural network architecture search over low-level primitives

Scale-permuted machine learning architecture

Training machine learning models using teacher annealing

Generating integrated circuit placements using neural networks

Contrastive pre-training for language tasks

Regularized neural network architecture search

Accurate and interpretable classification with hard attention

Performing machine learning tasks using instruction-tuned neural networks

Predicting a search engine ranking signal value

GOOGLE INC

Speech recognition with attention-based recurrent neural networks

Generating labeled images

Speech recognition with attention-based recurrent neural networks

Translating terms using numeric representations

Generating labeled images