Patents by Inventor Mostafa MAHMOUD
Mostafa MAHMOUD 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).
-
Patent number: 12299559Abstract: Described is a neural network accelerator tile. It includes an activation memory interface for interfacing with an activation memory to receive a set of activation representations and a weight memory interface for interfacing with a weight memory to receive a set of weight representations, and a processing element. The processing element is configured to implement a one-hot encoder, a histogrammer, an aligner, a reducer, and an accumulation sub-element which process the set of activation representations and the set of weight representations to produce a set of output representations.Type: GrantFiled: April 25, 2019Date of Patent: May 13, 2025Assignee: Samsung Electronics Co., Ltd.Inventors: Andreas Moshovos, Mostafa Mahmoud, Sayeh Sharifymoghaddam
-
Publication number: 20250039074Abstract: The present application relates to network routing between networks where private routing tables are in use and remote networks. Overlapping IP address ranges may be in use in private networks, which may result IP clashes in the remote network. A system (400) for routing internet protocol, IP, traffic is provided. The system (400) comprises one or more consumer networks (410) each comprising a gateway (420) and one or more private networks (430). Each of the plurality of private networks serves one or more endpoints (440). The system (400) further comprises a remote network (450) comprising a router (460) for each of the plurality of private networks (430) in the consumer network (410).Type: ApplicationFiled: October 21, 2022Publication date: January 30, 2025Inventors: Shifeeq AHMED, Mostafa Mahmoud Mohammed ELTAHER, Moamen Yehia AHMED, Stanislaw JAKIEL
-
Patent number: 12182621Abstract: A system and method for using sparsity to accelerate deep learning networks. The method includes: communicating a bit vector to a scheduler identifying which values in an input tensor are non-zero; for each lane of the input tensor, determining which values are to be communicated for multiply-accumulate (MAC) operations, the determination including directing performance of one of: communicating the current value in the lane; communicating the next value in the same lane where such value is non-zero; communicating a value from a step ahead in time where such value is non-zero; and communicating a value from a neighboring lane where such value is non-zero; and outputting the values of the MAC operations.Type: GrantFiled: July 16, 2021Date of Patent: December 31, 2024Assignee: THE GOVERNING COUNCIL OF THE UNIVERSITY OF TORONTOInventors: Mostafa Mahmoud, Andreas Moshovos
-
Patent number: 11836971Abstract: A processor-implemented method implementing a convolution neural network includes: determining a plurality of differential groups by grouping a plurality of raw windows of an input feature map into the plurality of differential groups; determining differential windows by performing, for each respective differential group of the differential groups, a differential operation between the raw windows of the respective differential group; determining a reference element of an output feature map corresponding to a reference raw window among the raw windows by performing a convolution operation between a kernel and the reference raw window; and determining remaining elements of the output feature map by performing a reference element summation operation based on the reference element and each of a plurality of convolution operation results determined by performing respective convolution operations between the kernel and each of the differential windows.Type: GrantFiled: August 23, 2019Date of Patent: December 5, 2023Assignee: Samsung Electronics Co., Ltd.Inventors: Mostafa Mahmoud, Andreas Moshovos
-
Publication number: 20230297337Abstract: A system and method for accelerating multiply-accumulate (MAC) floating-point units during training of deep learning networks. The method including: receiving a first input data stream A and a second input data stream B; adding exponents of the first data stream A and the second data stream B in pairs to produce product exponents; determining a maximum exponent using a comparator; determining a number of bits by which each significand in the second data stream has to be shifted prior to accumulation by adding product exponent deltas to the corresponding term in the first data stream and using an adder tree to reduce the operands in the second data stream into a single partial sum; adding the partial sum to a corresponding aligned value using the maximum exponent to determine accumulated values; and outputting the accumulated values.Type: ApplicationFiled: July 19, 2021Publication date: September 21, 2023Inventors: Omar MOHAMED AWAD, Mostafa MAHMOUD, Andreas MOSHOVOS
-
Publication number: 20230273828Abstract: A system and method for using sparsity to accelerate deep learning networks. The method includes: communicating a bit vector to a scheduler identifying which values in an input tensor are non-zero; for each lane of the input tensor, determining which values are to be communicated for multiply-accumulate (MAC) operations, the determination including directing performance of one of: communicating the current value in the lane; communicating the next value in the same lane where such value is non-zero; communicating a value from a step ahead in time where such value is non-zero; and communicating a value from a neighboring lane where such value is non-zero; and outputting the values of the MAC operations.Type: ApplicationFiled: July 16, 2021Publication date: August 31, 2023Inventors: Mostafa MAHMOUD, Andreas MOSHOVOS
-
Publication number: 20220305093Abstract: Natural killer (NK)-lysin peptides from chicken, bovine and human species are used as an antimicrobial or antiviral agent alone or in combination with other agents. A pharmaceutical composition comprising NK-lysin peptides was effective against Rotavirus and various gram-negative and gram-positive bacteria that were tested. Combinations of NK-lysin peptides plus silver nanoparticles or IL-2 were also effective against Salmonella typhimurium. Treatment with the NK-lysin peptides repressed ?-lactam-resistance genes (CTX-M-1, CTX-M-8 and CTX-M-9) in bacterial species tested and thus can be used to augment treatment with ?-lactams and increase bacterial sensitivity to this important class of antibiotics.Type: ApplicationFiled: January 12, 2022Publication date: September 29, 2022Inventors: Maged Mostafa MAHMOUD, Haitham A. YACOUB, Ahmed M. AL-HEJIN, Turki ABUJAML, Sherif ABD-ELMAKSOUD
-
Patent number: 11433122Abstract: Natural killer (NK)-lysin peptides from chicken, bovine and human species are used as an antimicrobial or antiviral agent alone or in combination with other agents. A pharmaceutical composition comprising NK-lysin peptides was effective against Rotavirus and various gram-negative and gram-positive bacteria that were tested. Combinations of NK-lysin peptides plus silver nanoparticles or IL-2 were also effective against Salmonella typhimurium. Treatment with the NK-lysin peptides repressed ?-lactam-resistance genes (CTX-M-1, CTX-M-8 and CTX-M-9) in bacterial species tested and thus can be used to augment treatment with ?-lactams and increase bacterial sensitivity to this important class of antibiotics.Type: GrantFiled: January 12, 2022Date of Patent: September 6, 2022Assignee: KING ABDULAZIZ UNIVERSITYInventors: Maged Mostafa Mahmoud, Haitham A. Yacoub, Ahmed M. Al-Hejin, Turki Abujaml, Sherif Abd-Elmaksoud
-
Patent number: 11425163Abstract: Methods, systems, and computer readable media to control a cyber physical system using an observer-based controller are described.Type: GrantFiled: February 5, 2020Date of Patent: August 23, 2022Assignee: King Fahd University of Petroleum and MineralsInventors: Magdi Sadek Mostafa Mahmoud, Mutaz M. Hamdan
-
Patent number: 11281232Abstract: Methods, computer readable media, and systems for systems and methods for multi-agent system control using consensus and saturation constraints are described.Type: GrantFiled: March 13, 2020Date of Patent: March 22, 2022Assignee: King Fahd University of Petroleum and MineralsInventors: Magdi Sadek Mostafa Mahmoud, Mojeed Opeyemi Oyedeji
-
Patent number: 11260113Abstract: Natural killer (NK)-lysin peptides from chicken, bovine and human species are used as an antimicrobial or antiviral agent alone or in combination with other agents. A pharmaceutical composition comprising NK-lysin peptides was effective against Rotavirus and various gram-negative and gram-positive bacteria that were tested. Combinations of NK-lysin peptides plus silver nanoparticles or IL-2 were also effective against Salmonella typhimurium. Treatment with the NK-lysin peptides repressed ?-lactam-resistance genes (CTX-M-1, CTX-M-8 and CTX-M-9) in bacterial species tested and thus can be used to augment treatment with ?-lactams and increase bacterial sensitivity to this important class of antibiotics.Type: GrantFiled: March 26, 2021Date of Patent: March 1, 2022Assignee: King Abdulaziz UniversityInventors: Maged Mostafa Mahmoud, Haitham A. Yacoub, Ahmed M. Al-Hejin, Turki Abujaml, Sherif Abd-Elmaksoud
-
Publication number: 20210286375Abstract: Methods, computer readable media, and systems for systems and methods for multi-agent system control using consensus and saturation constraints are described.Type: ApplicationFiled: March 13, 2020Publication date: September 16, 2021Applicant: King Fahd University of Petroleum and MineralsInventors: Magdi Sadek Mostafa MAHMOUD, Mojeed Opeyemi OYEDEJI
-
Publication number: 20210243224Abstract: Methods, systems, and computer readable media to control a cyber physical system using an observer-based controller are described.Type: ApplicationFiled: February 5, 2020Publication date: August 5, 2021Applicant: King Fahd University of Petroleum and MineralsInventors: Magdi Sadek Mostafa Mahmoud, Mutaz M. Hamdan
-
Publication number: 20210125046Abstract: Described is a neural network accelerator tile. It includes an activation memory interface for interfacing with an activation memory to receive a set of activation representations and a weight memory interface for interfacing with a weight memory to receive a set of weight representations, and a processing element. The processing element is configured to implement a one-hot encoder, a histogrammer, an aligner, a reducer, and an accumulation sub-element which process the set of activation representations and the set of weight representations to produce a set of output representations.Type: ApplicationFiled: April 25, 2019Publication date: April 29, 2021Inventors: Andreas Moshovos, Mostafa Mahmoud, Sayeh Sharifymoghaddam
-
Publication number: 20210004668Abstract: Described is a neural network accelerator tile for exploiting input sparsity. The tile includes a weight memory to supply each weight lane with a weight and a weight selection metadata, an activation selection unit to receive a set of input activation values and rearrange the set of input activation values to supply each activation lane with a set of rearranged activation values, a set of multiplexers including at least one multiplexer per pair of activation and weight lanes, where each multiplexer is configured to select a combination activation value for the activation lane from the activation lane set of rearranged activation values based on the weight lane weight selection metadata, and a set of combination units including at least one combination unit per multiplexer, where each combination unit is configured to combine the activation lane combination value with the weight lane weight to output a weight lane product.Type: ApplicationFiled: February 15, 2019Publication date: January 7, 2021Inventors: Andreas Moshovos, Alberto Delmas Lascorz, Zisis Poulos, Dylan Malone Stuart, Patrick Judd, Sayeh Sharify, Mostafa Mahmoud, Milos Nikolic, Kevin Chong Man Siu, Jorge Albericio
-
Publication number: 20200065646Abstract: A processor-implemented method implementing a convolution neural network includes: determining a plurality of differential groups by grouping a plurality of raw windows of an input feature map into the plurality of differential groups; determining differential windows by performing, for each respective differential group of the differential groups, a differential operation between the raw windows of the respective differential group; determining a reference element of an output feature map corresponding to a reference raw window among the raw windows by performing a convolution operation between a kernel and the reference raw window; and determining remaining elements of the output feature map by performing a reference element summation operation based on the reference element and each of a plurality of convolution operation results determined by performing respective convolution operations between the kernel and each of the differential windows.Type: ApplicationFiled: August 23, 2019Publication date: February 27, 2020Applicants: Samsung Electronics Co., Ltd., The Governing Council of the University of TorontoInventors: Mostafa MAHMOUD, Andreas MOSHOVOS
-
Patent number: 8226235Abstract: Apparatus and methods for measuring a dynamic deformation characteristic of a deformable target surface during a deformation interval. The measurement principles may be applied to a large variety of organic and inorganic materials having a surface that can be deformed by an applied non-contact force. The surface may be light diffusing and non-transparent or non-diffusing and transparent. A device for measuring a dynamic deformation characteristic of an in-vivo cornea during a deformation interval includes a corneal topographer and an air puff generator that are operationally integrated. Use of the inventive device enables a method for making a measurement of a deformation characteristic of the in-vivo cornea during a deformation interval, further allowing a determination of spatially-resolved in-vivo corneal biomechanical and biodynamic properties such as corneal elasticity and corneal viscosity.Type: GrantFiled: March 4, 2010Date of Patent: July 24, 2012Assignee: Vision Optimization, LLCInventors: Cynthia J. Roberts, James Richard Marous, Ashraf Mostafa Mahmoud
-
Publication number: 20100238408Abstract: Apparatus and methods for measuring a dynamic deformation characteristic of a deformable target surface during a deformation interval. The measurement principles may be applied to a large variety of organic and inorganic materials having a surface that can be deformed by an applied non-contact force. The surface may be light diffusing and non-transparent or non-diffusing and transparent. A device for measuring a dynamic deformation characteristic of an in-vivo cornea during a deformation interval includes a corneal topographer and an air puff generator that are operationally integrated. Use of the inventive device enables a method for making a measurement of a deformation characteristic of the in-vivo cornea during a deformation interval, further allowing a determination of spatially-resolved in-vivo corneal biomechanical and biodynamic properties such as corneal elasticity and corneal viscosity.Type: ApplicationFiled: March 4, 2010Publication date: September 23, 2010Applicant: VISION OPTIMIZATION, LLCInventors: Cynthia J. Roberts, James Richard Marous, Ashraf Mostafa Mahmoud