公开(公告)号：US20240311153A1

公开(公告)日：2024-09-19

申请号：US18331846

申请日：2023-06-08

Applicant: Microsoft Technology Licensing, LLC

Inventor： Behnaz ARZANI ,  Siva Kesava Reddy KAKARLA ,  Miguel OOM TEMUDO DE CASTRO ,  Srikanth KANDULA ,  Saeed MALEKI ,  Luke Jonathon MARSHALL

IPC: G06F9/30 ,  G06F9/38

CPC classification number: G06F9/3005 ,  G06F9/3877

Abstract: A method for scheduling a coordinated transfer of data among a plurality of processor nodes on a network comprises operating a multi-commodity flow model subject to a plurality of predetermined constraints. The model is configured to (a) receive as input a set of demands defining, for each of the plurality of processor nodes, an amount of data to be transferred to that processor node, (b) assign a plurality of paths linking the plurality of processor nodes, and (c) emit a schedule for transfer of the data along the plurality of paths so as to minimize a predetermined cost function, wherein the schedule comprises at least one store-and-forward operation and at least one copy operation.

公开(公告)号：US20180367550A1

公开(公告)日：2018-12-20

申请号：US15624614

申请日：2017-06-15

Applicant: Microsoft Technology Licensing, LLC

Inventor： Madanlal S. MUSUVATHI ,  Todd D. MYTKOWICZ ,  Saeed MALEKI ,  Yufei DING

IPC: H04L29/06 ,  G06N99/00 ,  G06N7/00

CPC classification number: H04L63/1416 ,  G06N7/005 ,  G06N20/00 ,  G06N20/20 ,  H04L63/0236 ,  H04L63/1425 ,  H04L63/1458

Abstract: Described herein is a system transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations (e.g., different instances of computing infrastructure) along with symbolic representations. Network transmission of the local models and the symbolic representations, rather than transmission of the large training data subsets processed to compute the local models and symbolic representations, conserves resources and decreases latency. The global model can then be used as a model to determine a likelihood that at least a portion of current and/or recently received data traffic is illegitimate data traffic that is associated with a cyber attack. In some instances, the system can implement a remedial action to mitigate the effects of the cyber attack on computing infrastructure.

公开(公告)号：US20180365580A1

公开(公告)日：2018-12-20

申请号：US15624555

申请日：2017-06-15

Applicant: Microsoft Technology Licensing, LLC

Inventor： Madanlal S. MUSUVATHI ,  Todd D. MYTKOWICZ ,  Saeed MALEKI ,  Yufei DING

IPC: G06N7/00 ,  G06N99/00 ,  G06F17/30

Abstract: Described herein is a system that transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations along with symbolic representations. The symbolic representations can be used to combine the local models. The global model can determine a likelihood, given a new data instance of a feature set, that a user performs a computer interaction with the content element. For instance, the system can use the model to provide search results in response to a search query submitted by a user. Or, the system can use the model to make a recommendation or suggestion to a user in response to a request for content (e.g., display a targeted advertisement, suggest a news story, etc.).

公开(公告)号：US20180365093A1

公开(公告)日：2018-12-20

申请号：US15624660

申请日：2017-06-15

Applicant: Microsoft Technology Licensing, LLC

Inventor： Madanlal S. MUSUVATHI ,  Todd D. MYTKOWICZ ,  Saeed MALEKI ,  Yufei DING

IPC: G06F11/07 ,  G06N99/00

Abstract: Described herein is a system that transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations along with symbolic representations. Network transmission of the local models and the symbolic representations, rather than transmission of the large training data subsets processed to compute the local models and symbolic representations, conserves resources and decreases latency. The global model can then be used as a model to determine a likelihood of a monitored resource or a user of the monitored resource experiencing a problem with respect to performance or completion of one or more operations. The system can also implement an action to assist in resolving or avoiding the problem.

公开(公告)号：US20240403598A1

公开(公告)日：2024-12-05

申请号：US18327821

申请日：2023-06-01

Applicant: Microsoft Technology Licensing, LLC

Inventor： Youshan MIAO ,  Fan YANG ,  Quanlu ZHANG ,  Saeed MALEKI ,  Xu CAO ,  Yi ZHU ,  Mao YANG ,  Lidong ZHOU ,  Zhiqi LIN

IPC: G06N3/04

Abstract: Embodiments of the present disclosure include techniques for designing and generating a parallelization plan for a neural network so that workloads in the neural network may be split amongst multiple devices. Operators and tensors in the neural network are transformed into a set of functionally equivalent operators and tensors. These functionally equivalent operators and tensors are then scheduled to separate devices for execution.

公开(公告)号：US20200076570A1

公开(公告)日：2020-03-05

申请号：US16177181

申请日：2018-10-31

Applicant: MICROSOFT TECHNOLOGY LICENSING, LLC

Inventor： Madanlal S. MUSUVATHI ,  Kim LAINE ,  Kristin E. LAUTER ,  Hao CHEN ,  Olli Ilari SAARIKIVI ,  Saeed MALEKI ,  Roshan DATHATHRI ,  Todd D. MYTKOWICZ

IPC: H04L9/00 ,  G06F21/72 ,  H04L12/26 ,  G06F21/60

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media directed to optimizing the generation, evaluation, and selection of tensor circuit specifications for a tensor circuit to perform homomorphic encryption operations on encrypted data. A computing device having an improved compiler and runtime configuration can obtain a tensor circuit and associated schema. The computing device can map the obtained tensor circuit to an equivalent tensor circuit, adapted to perform fully homomorphic encryption (FHE) operations, and instantiated based on the obtained associated scheme. The computing device can then monitor a flow of data through the equivalent FHE-adapted tensor circuit utilizing various tensor circuit specifications determined therefor. A cost of each tensor circuit specification can be determined by the computing device based on the monitored flow of data, so as to identify an optimal set of optimal tensor circuit specifications that can be employed by the obtained tensor circuit, to efficiently perform homomorphic encryption operations on encrypted data.

公开(公告)号：US20180365582A1

公开(公告)日：2018-12-20

申请号：US15624642

申请日：2017-06-15

Applicant: Microsoft Technology Licensing, LLC

Inventor： Madanlal S. MUSUVATHI ,  Todd D. MYTKOWICZ ,  Saeed MALEKI ,  Yufei DING

IPC: G06N7/08 ,  G06N99/00

Abstract: Described herein is a system that transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations along with symbolic representations. Network transmission of the local models and the symbolic representations, rather than transmission of the large training data subsets processed to compute the local models and symbolic representations, conserves resources and decreases latency. The global model can then be used as a model to determine a likelihood of a course of action being successful for an organization. For example, the course of action can be a purchase of a security or a business operation strategy. In another example, the course of action can be a type of medical treatment for a patient.

公开(公告)号：US20180330271A1

公开(公告)日：2018-11-15

申请号：US15600838

申请日：2017-05-22

Applicant: Microsoft Technology Licensing, LLC

Inventor： Saeed MALEKI ,  Madanlal S. MUSUVATHI ,  Todd D. MYTKOWICZ

IPC: G06N99/00 ,  G06N3/08 ,  G06N3/04

CPC classification number: G06N99/005 ,  G06K2009/485 ,  G06N3/0472 ,  G06N3/08 ,  G06N3/082 ,  G06N3/086

Abstract: Systems, methods, and computer-readable media are disclosed for parallel stochastic gradient descent using linear and non-linear activation functions. One method includes: receiving a set of input examples; receiving a global model; and learning a new global model based on the global model and the set of input examples by iteratively performing the following steps: computing a plurality of local models having a plurality of model parameters based on the global model and at least a portion of the set of input examples; computing, for each local model, a corresponding model combiner based on the global model and at least a portion of the set of input examples; and combining the plurality of local models into the new global model based on the current global model and the plurality of corresponding model combiners.