Inventor

VINYALS ORIOL

GB84 patents

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

VINYALS ORIOL

GOOGLE LLC

Training recurrent neural networks to generate sequences

Generating representations of input sequences using neural networks

Topic-based sequence modeling neural networks

Classifying videos using neural networks

Sentence compression using recurrent neural networks

Generating parse trees of text segments using neural networks

Speech recognition with attention-based recurrent neural networks

Generating parse trees of text segments using neural networks

Training distilled machine learning models

Generating target sequences from input sequences using partial conditioning

Neural machine translation systems with rare word processing

Universal transformers

Neural machine translation systems with rare word processing

Generating natural language descriptions of images

Convolutional, long short-term memory, fully connected deep neural networks

Processing and generating sets using recurrent neural networks

Generating target sequences from input sequences using partial conditioning

Speech recognition with attention-based recurrent neural networks

Convolutional, long short-term memory, fully connected deep neural networks

Training distilled machine learning models

Generating target sequences from input sequences using partial conditioning

Generating natural language descriptions of images

Generating output sequences from input sequences using neural networks

Universal transformers

Classifying videos using neural networks

DEEPMIND TECH LTD

Generating audio using neural networks

Processing sequences using convolutional neural networks

Multi-agent reinforcement learning with matchmaking policies

Auto-regressive neural network systems with a soft attention mechanism using support data patches

Learning observation representations by predicting the future in latent space

Generating audio using neural networks

Speech coding using content latent embedding vectors and speaker latent embedding vectors

Speech recognition using convolutional neural networks

Committed information rate variational autoencoders

Generating discrete latent representations of input data items

Feedforward generative neural networks

Multi-agent reinforcement learning with matchmaking policies

Generating audio using neural networks

Processing sequences using convolutional neural networks

Using hierarchical representations for neural network architecture searching

Graph neural network systems for generating structured representations of objects

Classifying input examples using a comparison set

Learning observation representations by predicting the future in latent space

Deep neural network system for similarity-based graph representations

Action selection neural network training using imitation learning in latent space

Sample-efficient adaptive text-to-speech

GOOGLE INC

Speech recognition with attention-based recurrent neural networks

Speech recognition with attention-based recurrent neural networks

Keyword detection without decoding

Generating natural language descriptions of images