Fine-grain compute communication execution for deep learning frameworks
S Sridharan, D Mudigere - US Patent App. 15/869,502, 2018 - Google Patents
One embodiment provides for a system to configure distributed training of a neural network.
The system includes memory to store a library to facilitate transmission of data during …
The system includes memory to store a library to facilitate transmission of data during …
Hardware implemented point to point communication primitives for machine learning
S Sridharan, K Vaidyanathan, D Das - US Patent 11,488,008, 2022 - Google Patents
One embodiment provides for a system to compute and distribute data for distributed training
of a neural network, the system including first memory to store a first set of instructions …
of a neural network, the system including first memory to store a first set of instructions …
Communication optimizations for distributed machine learning
S Sridharan, K Vaidyanathan, D Das… - US Patent …, 2022 - Google Patents
Embodiments described herein provide a system to config ure distributed training of a neural
network, the system comprising memory to store a library to facilitate data transmission …
network, the system comprising memory to store a library to facilitate data transmission …
Deep Learning Training System
TA Chilimbi, Y Suzue, JR Apacible… - US Patent App. 14 …, 2015 - Google Patents
Training large neural network models by providing training input to model training machines
organized as multiple replicas that asynchronously update a shared model via a global …
organized as multiple replicas that asynchronously update a shared model via a global …
Memory efficient scalable deep learning with model parallelization
Methods and systems for training a neural network include sampling multiple local sub-
networks from a global neural network. The local sub-networks include a subset of neurons …
networks from a global neural network. The local sub-networks include a subset of neurons …
Tool for facilitating efficiency in machine learning
R Barik, BT Lewis, M Sundaresan, J Jackson… - US Patent …, 2022 - Google Patents
A mechanism is described for facilitating smart distribution of resources for deep learning
autonomous machines. A method of embodiments, as described herein, includes detecting …
autonomous machines. A method of embodiments, as described herein, includes detecting …
Training neural networks represented as computational graphs
Y Yu, MK Venkatakrishna - US Patent 10,970,628, 2021 - Google Patents
Systems and Methods for training a neural network repre sented as a computational graph
are disclosed. An example method begins with obtaining data representing a computa tional …
are disclosed. An example method begins with obtaining data representing a computa tional …
Hardware accelerated neural network subgraphs
RK Kovvuri, AM El Husseini, SK Reinhardt… - US Patent App. 15 …, 2019 - Google Patents
Technology related to hardware accelerated neural network subgraphs is disclosed. In one
example of the disclosed technology, a method includes receiving source code specifying a …
example of the disclosed technology, a method includes receiving source code specifying a …
Sparse inference modules for deep learning
Described is a sparse inference module that can be incorporated into a deep learning
system. For example, the deep learning system includes a plurality of hierarchical feature …
system. For example, the deep learning system includes a plurality of hierarchical feature …
Compute optimization mechanism for deep neural networks
US20190205736A1 - Compute optimization mechanism for deep neural networks - Google
Patents US20190205736A1 - Compute optimization mechanism for deep neural networks …
Patents US20190205736A1 - Compute optimization mechanism for deep neural networks …