Inventor · disambiguated record
Ali Shafiee Ardestani
Also filed as: SHAFIEE ARDESTANI ALI
35 granted patents·10 pending applications·44 citations·filing 2016–2025
95Inventor score
Top patents by PatentIndex Score
45 records- 0195US11507817B2System and method for performing computations for deep neural networksSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Nov 22, 2022·7 cites·18 claims
- 0294US11687764B2System and method for increasing utilization of dot-product based neural network acceleratorSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Jun 27, 2023·4 cites·20 claims
- 0392US10055383B1Matrix circuitsHEWLETT PACKARD ENTPR DEV LP·Filed 2017·Granted Aug 21, 2018·11 cites·20 claims
- 0490US10942673B2Data processing using resistive memory arraysHEWLETT PACKARD ENTPR DEV LP·Filed 2016·Granted Mar 9, 2021·11 cites·18 claims
- 0587US11775611B2Piecewise quantization for neural networksSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Oct 3, 2023·3 cites·20 claims
- 0685US12314833B2Neural processorSAMSUNG ELECTRONICS CO LTD·Filed 2024·Granted May 27, 2025·0 cites·20 claims
- 0783US11880760B2Mixed-precision NPU tile with depth-wise convolutionSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Jan 23, 2024·2 cites·18 claims
- 0883US2025232152A1Neural processorSAMSUNG ELECTRONICS CO LTD·Filed 2025·Application pending·0 cites
- 0982US12073302B2Neural processorSAMSUNG ELECTRONICS CO LTD·Filed 2023·Granted Aug 27, 2024·0 cites·19 claims
- 1081US10664271B2Dot product engine with negation indicatorHEWLETT PACKARD ENTPR DEV LP·Filed 2016·Granted May 26, 2020·3 cites·20 claims
- 1177US2025028505A1Accelerating 2d convolutional layer mapping on a dot product architectureSAMSUNG ELECTRONICS CO LTD·Filed 2024·Application pending·0 cites
- 1276US12136031B2System and method for increasing utilization of dot-product based neural network acceleratorSAMSUNG ELECTRONICS CO LTD·Filed 2023·Granted Nov 5, 2024·0 cites·19 claims
- 1376US12015429B2Hardware channel-parallel data compression/decompressionSAMSUNG ELECTRONICS CO LTD·Filed 2022·Granted Jun 18, 2024·0 cites·20 claims
- 1476US11775801B2Neural processorSAMSUNG ELECTRONICS CO LTD·Filed 2019·Granted Oct 3, 2023·1 cites·20 claims
- 1576US11775256B2Signed multiplication using unsigned multiplier with dynamic fine-grained operand isolationSAMSUNG ELECTRONICS CO LTD·Filed 2023·Granted Oct 3, 2023·0 cites·20 claims
- 1676US11775802B2Neural processorSAMSUNG ELECTRONICS CO LTD·Filed 2019·Granted Oct 3, 2023·1 cites·20 claims
- 1775US10963220B2Signed multiplication using unsigned multiplier with dynamic fine-grained operand isolationSAMSUNG ELECTRONICS CO LTD·Filed 2019·Granted Mar 30, 2021·1 cites·22 claims
- 1869US11579842B2Signed multiplication using unsigned multiplier with dynamic fine-grained operand isolationSAMSUNG ELECTRONICS CO LTD·Filed 2021·Granted Feb 14, 2023·0 cites·23 claims
- 1968US11671111B2Hardware channel-parallel data compression/decompressionSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Jun 6, 2023·0 cites·20 claims
- 2067US11681907B2System and method for performing computations for deep neural networksSAMSUNG ELECTRONICS CO LTD·Filed 2022·Granted Jun 20, 2023·0 cites·18 claims
- 2166US2024095519A1Extreme sparse deep learning edge inference acceleratorSAMSUNG ELECTRONICS CO LTD·Filed 2022·Application pending·0 cites
- 2264US12086700B2Neural processorSAMSUNG ELECTRONICS CO LTD·Filed 2019·Granted Sep 10, 2024·0 cites·20 claims
- 2361US12299960B2Efficiency of vision transformers with adaptive token pruningSAMSUNG ELECTRONICS CO LTD·Filed 2022·Granted May 13, 2025·0 cites·20 claims
- 2461US12112141B2Accelerating 2D convolutional layer mapping on a dot product architectureSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Oct 8, 2024·0 cites·18 claims
- 2555US12400108B2Mixed-precision neural network accelerator tile with lattice fusionSAMSUNG ELECTRONICS CO LTD·Filed 2021·Granted Aug 26, 2025·0 cites·18 claims
- 2655US2022156568A1Dual-sparse neural processing unit with multi-dimensional routing of non-zero valuesSAMSUNG ELECTRONICS CO LTD·Filed 2021·Application pending·0 cites
- 2754US2022413805A1Partial sum compressionSAMSUNG ELECTRONICS CO LTD·Filed 2021·Application pending·0 cites
- 2854US2024192922A1System and method for handling processing with sparse weights and outliersSAMSUNG ELECTRONICS CO LTD·Filed 2023·Application pending·0 cites
- 2952US12361266B2Hierarchical weight preprocessing for neural network acceleratorSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Jul 15, 2025·0 cites·20 claims
- 3051US12423566B2SRAM-sharing for reconfigurable neural processing unitsSAMSUNG ELECTRONICS CO LTD·Filed 2021·Granted Sep 23, 2025·0 cites·17 claims
- 3151US2023103997A1Integrating spatial locality into image transformers with masked attentionSAMSUNG ELECTRONICS CO LTD·Filed 2022·Application pending·0 cites
- 3250US12229659B2Processor with outlier accommodationSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Feb 18, 2025·0 cites·20 claims
- 3350US12158923B2Low overhead implementation of Winograd for CNN with 3x3, 1x3 and 3x1 filters on weight station dot-product based CNN acceleratorsSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Dec 3, 2024·0 cites·12 claims
- 3450US12001929B2Mixed-precision neural processing unit (NPU) using spatial fusion with load balancingSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Jun 4, 2024·0 cites·20 claims
- 3548US10090030B1Signal conversion using an analog-to-digital converter and reference voltage comparisonHEWLETT PACKARD ENTPR DEV LP·Filed 2017·Granted Oct 2, 2018·0 cites·19 claims
- 3648US2023153586A1Accelerate neural networks with compression at different levelsSAMSUNG ELECTRONICS CO LTD·Filed 2022·Application pending·0 cites
- 3747US2023205488A1Efficient circuit for neural network processingSAMSUNG ELECTRONICS CO LTD·Filed 2022·Application pending·0 cites
- 3846US12216735B2Supporting floating point 16 (FP16) in dot product architectureSAMSUNG ELECTRONICS CO LTD·Filed 2021·Granted Feb 4, 2025·0 cites·19 claims
- 3946US11861327B2Processor for fine-grain sparse integer and floating-point operationsSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Jan 2, 2024·0 cites·20 claims
- 4046US11126549B2Processing in-memory architectures for performing logical operationsHEWLETT PACKARD ENTPR DEV LP·Filed 2016·Granted Sep 21, 2021·0 cites·20 claims
- 4145US11861328B2Processor for fine-grain sparse integer and floating-point operationsSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Jan 2, 2024·0 cites·20 claims
- 4245US2018341623A1Matrix circuitsHEWLETT PACKARD ENTPR DEV LP·Filed 2018·Application pending·0 cites
- 4344US12182577B2Neural-processing unit tile for shuffling queued nibbles for multiplication with non-zero weight nibblesSAMSUNG ELECTRONICS CO LTD·Filed 2020·Granted Dec 31, 2024·0 cites·19 claims
- 4442US10754582B2Assigning data to a resistive memory array based on a significance levelHEWLETT PACKARD ENTPR DEV LP·Filed 2016·Granted Aug 25, 2020·0 cites·15 claims
- 4540US10754581B2Identifying outlying values in matricesHEWLETT PACKARD ENTPR DEV LP·Filed 2016·Granted Aug 25, 2020·0 cites·15 claims
Join the waitlist — get patent alerts
Get an alert when Ali Shafiee Ardestani files or is granted a new patent.
We store only your email — no account needed. See our privacy policy.
Identity basis: PatentsView inventor disambiguation (2025Q4-odp release). How scoring works →