Inventor · disambiguated record
Abdulkadir Utku Diril
Also filed as: DIRIL ABDULKADIR U · DIRIL ABDULKADIR UTKU
45 granted patents·8 pending applications·208 citations·filing 2006–2023
97Inventor score
Top patents by PatentIndex Score
53 records- 0195US11138292B1Circuit and method for computing depthwise convolutionFACEBOOK INC·Filed 2019·Granted Oct 5, 2021·35 cites·20 claims
- 0294US11520853B2Mapping convolution to a partition channel convolution engineMETA PLATFORMS INC·Filed 2020·Granted Dec 6, 2022·5 cites·20 claims
- 0392US11599181B1Systems and methods for reducing power consumption of convolution operations of artificial neural networksMETA PLATFORMS INC·Filed 2019·Granted Mar 7, 2023·10 cites·20 claims
- 0492US10699190B1Systems and methods for efficiently updating neural networksFACEBOOK INC·Filed 2018·Granted Jun 30, 2020·15 cites·16 claims
- 0591US11487888B1Systems and methods for protecting neural network weightsMETA PLATFORMS INC·Filed 2020·Granted Nov 1, 2022·3 cites·20 claims
- 0691US10948966B1Systems and methods for optimizing power usage for systems within quality-of-service constraintsFACEBOOK INC·Filed 2018·Granted Mar 16, 2021·11 cites·20 claims
- 0791US10817783B1Systems and methods for efficiently updating neural networksFACEBOOK INC·Filed 2020·Granted Oct 27, 2020·3 cites·20 claims
- 0890US11537865B2Mapping convolution to a channel convolution engineMETA PLATFORMS INC·Filed 2020·Granted Dec 27, 2022·3 cites·20 claims
- 0990US11443013B2Pipelined pointwise convolution using per-channel convolution operationsMETA PLATFORMS INC·Filed 2020·Granted Sep 13, 2022·3 cites·20 claims
- 1090US10671147B2Dynamic power management for artificial intelligence hardware acceleratorsFACEBOOK INC·Filed 2017·Granted Jun 2, 2020·7 cites·20 claims
- 1189US11468313B1Systems and methods for quantizing neural networks via periodic regularization functionsMETA PLATFORMS INC·Filed 2018·Granted Oct 11, 2022·18 cites·20 claims
- 1289US8060765B1Power estimation based on block activityCHA HUNGSE·Filed 2006·Granted Nov 15, 2011·41 cites·20 claims
- 1388US10255655B1Serial pixel processing with storage for overlapping texel dataAPPLE INC·Filed 2017·Granted Apr 9, 2019·7 cites·20 claims
- 1486US10579383B1Systems and methods for efficient scaling of quantized integersFACEBOOK INC·Filed 2018·Granted Mar 3, 2020·6 cites·20 claims
- 1584US11514306B1Static memory allocation in neural networksMETA PLATFORMS INC·Filed 2018·Granted Nov 29, 2022·4 cites·20 claims
- 1683US10834385B1Systems and methods for encoding videos using reference objectsFACEBOOK INC·Filed 2018·Granted Nov 10, 2020·3 cites·20 claims
- 1783US10719613B1Systems and methods for protecting neural network weightsFACEBOOK INC·Filed 2018·Granted Jul 21, 2020·5 cites·19 claims
- 1882US11501147B1Systems and methods for handling padding regions in convolution operationsMETA PLATFORMS INC·Filed 2020·Granted Nov 15, 2022·2 cites·20 claims
- 1981US11256977B2Lowering hardware for neural networksFACEBOOK INC·Filed 2017·Granted Feb 22, 2022·4 cites·20 claims
- 2081US10482156B2Sparsity-aware hardware acceleratorsFACEBOOK INC·Filed 2017·Granted Nov 19, 2019·4 cites·20 claims
- 2179US11054998B1High bandwidth memory system with distributed request broadcasting mastersFACEBOOK INC·Filed 2019·Granted Jul 6, 2021·3 cites·2 claims
- 2278US11481471B2Mapping convolution to a matrix processor unitMETA PLATFORMS INC·Filed 2019·Granted Oct 25, 2022·3 cites·20 claims
- 2378US10474430B2Mixed-precision processing elements, systems, and methods for computational modelsFACEBOOK INC·Filed 2017·Granted Nov 12, 2019·2 cites·19 claims
- 2478US10372787B2Hardware accelerator pre-configured with coefficients for matrix-transform operationsFACEBOOK INC·Filed 2017·Granted Aug 6, 2019·3 cites·20 claims
- 2576US11580192B2Grouped convolution using point-to-point connected channel convolution enginesMETA PLATFORMS INC·Filed 2020·Granted Feb 14, 2023·1 cites·20 claims
- 2676US11275560B2Hardware for floating-point arithmetic in multiple formatsMETA PLATFORMS INC·Filed 2020·Granted Mar 15, 2022·1 cites·20 claims
- 2776US10553207B2Systems and methods for employing predication in computational modelsFACEBOOK INC·Filed 2017·Granted Feb 4, 2020·3 cites·13 claims
- 2872US2023252263A1Point to point connected processing elements with data joiner componentsMETA PLATFORMS INC·Filed 2023·Application pending·0 cites
- 2970US2023251903A1High bandwidth memory system with dynamically programmable distribution schemeMETA PLATFORMS INC·Filed 2023·Application pending·0 cites
- 3067US11537301B2High bandwidth memory system with distributed request broadcasting mastersMETA PLATFORMS INC·Filed 2021·Granted Dec 27, 2022·0 cites·20 claims
- 3164US2023004624A1High throughput matrix processor with support for concurrently processing multiple matricesMETA PLATFORMS INC·Filed 2022·Application pending·0 cites
- 3262US11657252B2Point to point connected processing elements with data joiner componentsMETA PLATFORMS INC·Filed 2019·Granted May 23, 2023·0 cites·20 claims
- 3362US11264011B2Systems and methods for employing predication in computational modelsFACEBOOK INC·Filed 2020·Granted Mar 1, 2022·0 cites·20 claims
- 3462US8416241B2Efficient tile-based rasterizationDIRIL ABDULKADIR UTKU·Filed 2011·Granted Apr 9, 2013·2 cites·19 claims
- 3560US11023240B1Systems and methods for efficient scaling of quantized integersFACEBOOK INC·Filed 2019·Granted Jun 1, 2021·0 cites·20 claims
- 3658US11762560B2Optimizing NOC performance using crossbarsMETA PLATFORMS INC·Filed 2021·Granted Sep 19, 2023·0 cites·18 claims
- 3758US11663043B2High bandwidth memory system with dynamically programmable distribution schemeMETA PLATFORMS INC·Filed 2019·Granted May 30, 2023·0 cites·20 claims
- 3858US11409838B2High throughput matrix processor with support for concurrently processing multiple matricesMETA PLATFORMS INC·Filed 2019·Granted Aug 9, 2022·0 cites·20 claims
- 3957US2024348263A1Selective and flexible compression of data associated with a machine learning modelMETA PLATFORMS INC·Filed 2023·Application pending·0 cites
- 4056US8009169B2Efficient tile-based rasterizationVIVANTE CORP·Filed 2007·Granted Aug 30, 2011·1 cites·19 claims
- 4153US11699081B2Systems and methods for reducing data movement during convolution operations in artificial neural networksMETA PLATFORMS INC·Filed 2019·Granted Jul 11, 2023·0 cites·20 claims
- 4253US11688032B2Three-dimensional convolution pipeline with memory organizer unitMETA PLATFORMS INC·Filed 2019·Granted Jun 27, 2023·0 cites·20 claims
- 4351US11520854B2Support for different matrix multiplications by selecting adder tree intermediate resultsMETA PLATFORMS INC·Filed 2019·Granted Dec 6, 2022·0 cites·20 claims
- 4448US11531619B2High bandwidth memory system with crossbar switch for dynamically programmable distribution schemeMETA PLATFORMS INC·Filed 2019·Granted Dec 20, 2022·0 cites·20 claims
- 4548US2021334072A1Mapping convolution to connected processing elements using distributed pipelined separable convolution operationsFACEBOOK INC·Filed 2020·Application pending·0 cites
- 4645US11614920B2Bypassing zero-value multiplications in a hardware multiplierMETA PLATFORMS INC·Filed 2020·Granted Mar 28, 2023·0 cites·20 claims
- 4745US8553046B2Thin-line detection apparatus and methodZHONG LEFAN·Filed 2007·Granted Oct 8, 2013·0 cites·20 claims
- 4844US10354431B2Level of detail offset determinationAPPLE INC·Filed 2015·Granted Jul 16, 2019·0 cites·20 claims
- 4944US9600909B2Processed texel cacheAPPLE INC·Filed 2015·Granted Mar 21, 2017·0 cites·20 claims
- 5044US2020364047A1High throughput neural network operations using inter-layer memory layout transformationFACEBOOK INC·Filed 2019·Application pending·0 cites
Showing the top 50 of 53 patent records by PatentIndex Score.
Identity basis: PatentsView inventor disambiguation (2025Q4-odp release). How scoring works →