Patents by Inventor Hanxiao Liu
Hanxiao Liu has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20250139431Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing a machine learning task on a network input to generate a network output. In one aspect, one of the systems includes a neural network configured to perform the machine learning task, the neural network including one or more attentive layers that each include a gated attention unit.Type: ApplicationFiled: January 30, 2023Publication date: May 1, 2025Inventors: Hanxiao Liu, Weizhe Hua, Zihang Dai, Quoc V. Le
-
Publication number: 20250131251Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing a machine learning task on a network input to generate a network output. In one aspect, one of the systems includes a neural network configured to perform the machine learning task, the neural network including one or more expert neural network blocks that each include router that performs expert-choice routing between multiple expert neural networks.Type: ApplicationFiled: January 30, 2023Publication date: April 24, 2025Inventors: Hanxiao Liu, Quoc V. Le, Yanqi Zhou, Tao Lei, Yuzhe Zhao, Yanping Huang, Nan Du, Zhifeng Chen, Andrew M. Dai, James Laudon
-
Publication number: 20250124700Abstract: Methods, systems, and apparatus, including computer-readable media, are described for processing an input image using a convolutional neural network (CNN). The CNN includes a sequence of layer blocks. Each of a first subset of the layer blocks in the sequence is configured to perform operations that include: i) receiving an input feature map for the layer block, ii) generating an expanded feature map from the input feature map using a group convolution, and iii) generating a reduced feature map from the expanded feature map. The input feature map is an h w feature map with c1 channels. The expanded feature map is an h w feature map with c2 channels, whereas the reduced feature map is an h w feature map with c1 channels. C2 is greater than c1. An output feature map is generated for the layer block from the reduced feature map.Type: ApplicationFiled: October 8, 2021Publication date: April 17, 2025Inventors: Berkin Akin, Suyog Gupta, Cao Gao, Ping Zhou, Gabriel Mintzer Bender, Hanxiao Liu
-
Publication number: 20240428071Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing a machine learning task on a network input to generate a network output. One of the systems includes an attention neural network configured to perform the machine learning task. The attention neural network includes one or more attentions layers that each include a squared ReLU activation layer, a depth-wise convolution layer, or both.Type: ApplicationFiled: September 3, 2024Publication date: December 26, 2024Inventors: David Richard So, Quoc V. Le, Hanxiao Liu, Wojciech Andrzej Manke, Zihang Dai, Noam M. Shazeer
-
Publication number: 20240386260Abstract: Methods, systems, and apparatus, including computer-readable media, are described for processing an input image using integrated circuit that implements a convolutional neural network with a group convolution layer. The processing includes determining a mapping of partitions along a channel dimension of an input feature map to multiply accumulate cells (MACs) in a computational unit of the circuit and applying a group convolution to the input feature map. Applying the group convolution includes, for each partition: providing weights for the group convolution layer to a subset of MACs based on the mapping; providing, via an input bus of the circuit, an input of the feature map to each MAC in the subset; and computing, at each MAC in the subset, a product using the input and a weight for the group convolution layer. An output feature map is generated for the group convolution layer based on an accumulation of products.Type: ApplicationFiled: October 8, 2021Publication date: November 21, 2024Inventors: Berkin Akin, Suyog Gupta, Cao Gao, Ping Zhou, Gabriel Mintzer Bender, Hanxiao Liu
-
Publication number: 20240249146Abstract: A computer-implemented method for automatically determining a neural network architecture represents a neural network architecture as a data structure defining a hierarchical set of directed acyclic graphs in multiple levels. Each graph has an input, an output, and a plurality of nodes between the input and the output. At each level, a corresponding set of the nodes are connected pairwise by directed edges which indicate operations performed on outputs of one node to generate an input to another node. Each level is associated with a corresponding set of operations. At a lowest level, the operations associated with each edge are selected from a set of primitive operations. The method includes repeatedly generating new sample neural network architectures, and evaluating their fitness. The modification is performed by selecting a level, selecting two nodes at that level, and modifying, removing or adding an edge between those nodes according to operations associated with lower levels of the hierarchy.Type: ApplicationFiled: January 17, 2024Publication date: July 25, 2024Inventors: Chrisantha Thomas Fernando, Karen Simonyan, Koray Kavukcuoglu, Hanxiao Liu, Oriol Vinyals
-
Patent number: 11907853Abstract: A computer-implemented method for automatically determining a neural network architecture represents a neural network architecture as a data structure defining a hierarchical set of directed acyclic graphs in multiple levels. Each graph has an input, an output, and a plurality of nodes between the input and the output. At each level, a corresponding set of the nodes are connected pairwise by directed edges which indicate operations performed on outputs of one node to generate an input to another node. Each level is associated with a corresponding set of operations. At a lowest level, the operations associated with each edge are selected from a set of primitive operations. The method includes repeatedly generating new sample neural network architectures, and evaluating their fitness. The modification is performed by selecting a level, selecting two nodes at that level, and modifying, removing or adding an edge between those nodes according to operations associated with lower levels of the hierarchy.Type: GrantFiled: October 26, 2018Date of Patent: February 20, 2024Assignee: DeepMind Technologies LimitedInventors: Chrisantha Thomas Fernando, Karen Simonyan, Koray Kavukcuoglu, Hanxiao Liu, Oriol Vinyals
-
Publication number: 20230359862Abstract: A computer-implemented method for performing computer vision with reduced computational cost and improved accuracy can include obtaining, by a computing system including one or more computing devices, input data comprising an input tensor having one or more dimensions, providing, by the computing system, the input data to a machine-learned convolutional attention network, the machine-learned convolutional attention network including two or more network stages, and, in response to providing the input data to the machine-learned convolutional attention network, receiving, by the computing system, a machine-learning prediction from the machine-learned convolutional attention network. The convolutional attention network can include at least one attention block, wherein the attention block includes a relative attention mechanism, the relative attention mechanism including the sum of a static convolution kernel with an adaptive attention matrix.Type: ApplicationFiled: July 19, 2023Publication date: November 9, 2023Inventors: Zihang Dai, Mingxing Tan, Quoc V. Le, Hanxiao Liu
-
Patent number: 11755883Abstract: A computer-implemented method for performing computer vision with reduced computational cost and improved accuracy can include obtaining, by a computing system including one or more computing devices, input data comprising an input tensor having one or more dimensions, providing, by the computing system, the input data to a machine-learned convolutional attention network, the machine-learned convolutional attention network including two or more network stages, and, in response to providing the input data to the machine-learned convolutional attention network, receiving, by the computing system, a machine-learning prediction from the machine-learned convolutional attention network. The convolutional attention network can include at least one attention block, wherein the attention block includes a relative attention mechanism, the relative attention mechanism including the sum of a static convolution kernel with an adaptive attention matrix.Type: GrantFiled: May 27, 2022Date of Patent: September 12, 2023Assignee: GOOGLE LLCInventors: Zihang Dai, Hanxiao Liu, Mingxing Tan, Quoc V. Le
-
Publication number: 20230176840Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for compiler optimizations using a compiler optimization network. One of the methods includes receiving an input program, wherein the input program defines a graph of operation modules, wherein each node in the graph is a respective operation module, and each edge between nodes in the graph represents one operation module receiving the output generated by another operation module. The input program is processed by a compiler optimization network comprising a graph-embedding network that is configured to encode operation features and operation dependencies of the operation modules of the input program into a graph embedding representation and a policy network that is configured to generate an optimization action for each of one or more nodes encoded in the graph embedding representation.Type: ApplicationFiled: June 7, 2021Publication date: June 8, 2023Inventors: Yanqi Zhou, Sudip Roy, Amirali Abdolrashidi, Daniel Lin-Kit Wong, Chao Ma, Qiumin Xu, Hanxiao Liu, Phitchaya Mangpo Phothilimthana, Shen Wang, Anna Darling Goldie, Azalia Mirhoseini, James Laudon
-
Publication number: 20230154161Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using memory-optimized contrastive learning to train image encoder and text encoder neural networks.Type: ApplicationFiled: November 16, 2022Publication date: May 18, 2023Inventors: Hieu Hy Pham, Zihang Dai, Golnaz Ghiasi, Hanxiao Liu, Wei Yu, Mingxing Tan, Quoc V. Le
-
Publication number: 20230121404Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for searching for an architecture for an activation-normalization layer to be included in a neural network to replace a set of layers that receive a layer input comprising a plurality of values, apply one or more normalization operations to the values in the layer input to generate a normalized layer input, and apply an element-wise activation function to the normalized layer input to generate a layer output.Type: ApplicationFiled: February 8, 2021Publication date: April 20, 2023Inventors: Hanxiao Liu, Quoc V. Le, Andrew Brock, Karen Simonyan
-
Publication number: 20220405579Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for selecting a neural network to perform a particular machine learning task while satisfying a set of constraints.Type: ApplicationFiled: March 3, 2021Publication date: December 22, 2022Inventors: Jiahui Yu, Pengchong Jin, Hanxiao Liu, Gabriel Mintzer Bender, Pieter-Jan Kindermans, Mingxing Tan, Xiaodan Song, Ruoming Pang, Quoc V. Le
-
Publication number: 20220383119Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing a machine learning task on a network input to generate a network output. One of the systems includes an attention neural network configured to perform the machine learning task. The attention neural network includes one or more attentions layers that each include a squared ReLU activation layer, a depth-wise convolution layer, or both.Type: ApplicationFiled: May 27, 2022Publication date: December 1, 2022Inventors: David Richard So, Quoc V. Le, Jr., Hanxiao Liu, Wojciech Andrzej Manke, Zihang Dai, Noam M. Shazeer
-
Publication number: 20220383069Abstract: A computer-implemented method for performing computer vision with reduced computational cost and improved accuracy can include obtaining, by a computing system including one or more computing devices, input data comprising an input tensor having one or more dimensions, providing, by the computing system, the input data to a machine-learned convolutional attention network, the machine-learned convolutional attention network including two or more network stages, and, in response to providing the input data to the machine-learned convolutional attention network, receiving, by the computing system, a machine-learning prediction from the machine-learned convolutional attention network. The convolutional attention network can include at least one attention block, wherein the attention block includes a relative attention mechanism, the relative attention mechanism including the sum of a static convolution kernel with an adaptive attention matrix.Type: ApplicationFiled: May 27, 2022Publication date: December 1, 2022Inventors: Zihang Dai, Hanxiao Liu, Mingxing Tan, Quoc V. Le
-
Publication number: 20220367052Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing a machine learning task on a network input to generate a network output. In one aspect, one of the systems includes a neural network configured to perform the machine learning task, the neural network including one or more blocks that each include a feedforward spatial transformation unit.Type: ApplicationFiled: May 16, 2022Publication date: November 17, 2022Inventors: Hanxiao Liu, David Richard So, Quoc V. Le, Zihang Dai
-
Publication number: 20200293899Abstract: A computer-implemented method for automatically determining a neural network architecture represents a neural network architecture as a data structure defining a hierarchical set of directed acyclic graphs in multiple levels. Each graph has an input, an output, and a plurality of nodes between the input and the output. At each level, a corresponding set of the nodes are connected pairwise by directed edges which indicate operations performed on outputs of one node to generate an input to another node. Each level is associated with a corresponding set of operations. At a lowest level, the operations associated with each edge are selected from a set of primitive operations. The method includes repeatedly generating new sample neural network architectures, and evaluating their fitness. The modification is performed by selecting a level, selecting two nodes at that level, and modifying, removing or adding an edge between those nodes according to operations associated with lower levels of the hierarchy.Type: ApplicationFiled: October 26, 2018Publication date: September 17, 2020Inventors: Chrisantha Thomas Fernando, Karen Simonyan, Koray Kavukcuoglu, Hanxiao Liu, Oriol Vinyals