Inventor · disambiguated record
Arnab Raha
Also filed as: Raha Arnab
14 granted patents·38 pending applications·5 citations·filing 2016–2025
84Inventor score
Top patents by PatentIndex Score
52 records- 0190US12141683B2Performance scaling for dataflow deep neural network hardware acceleratorsINTEL CORP·Filed 2021·Granted Nov 12, 2024·3 cites·20 claims
- 0285US2025365009A1Methods, systems, articles of manufacture, and apparatus to decode zero-value-compression data vectorsINTEL CORP·Filed 2025·Application pending·0 cites
- 0382US12147836B2Schedule-aware dynamically reconfigurable adder tree architecture for partial sum accumulation in machine learning acceleratorsINTEL CORP·Filed 2021·Granted Nov 19, 2024·1 cites·31 claims
- 0481US11922178B2Methods and apparatus to load data within a machine learning acceleratorINTEL CORP·Filed 2021·Granted Mar 5, 2024·1 cites·20 claims
- 0578US12438553B2Methods, systems, articles of manufacture, and apparatus to decode zero-value-compression data vectorsINTEL CORP·Filed 2023·Granted Oct 7, 2025·0 cites·20 claims
- 0676US2025322218A1System and method for balancing sparsity in weights for accelerating deep neural networksINTEL CORP·Filed 2025·Application pending·0 cites
- 0775US12288153B2Schedule-aware tensor distribution moduleINTEL CORP·Filed 2024·Granted Apr 29, 2025·0 cites·20 claims
- 0875US12242861B2Methods and apparatus to load data within a machine learning acceleratorINTEL CORP·Filed 2024·Granted Mar 4, 2025·0 cites·17 claims
- 0974US2025036928A1Performance scaling for dataflow deep neural network hardware acceleratorsINTEL CORP·Filed 2024·Application pending·0 cites
- 1073US2025028565A1Schedule-aware dynamically reconfigurable adder tree architecture for partial sum accumulation in machine learning acceleratorsINTEL CORP·Filed 2024·Application pending·0 cites
- 1168US2025377861A1Energy-efficient pre-encoded booth for stationary weights and activationsINTEL CORP·Filed 2025·Application pending·0 cites
- 1265US12367380B2System and method for balancing sparsity in weights for accelerating deep neural networksINTEL CORP·Filed 2021·Granted Jul 22, 2025·0 cites·25 claims
- 1365US11804851B2Methods, systems, articles of manufacture, and apparatus to decode zero-value-compression data vectorsINTEL CORP·Filed 2020·Granted Oct 31, 2023·0 cites·21 claims
- 1464US2025060940A1Neural network accelerator performing operation with mixed-format weightsINTEL CORP·Filed 2024·Application pending·0 cites
- 1563US2025307656A1Graph neural network execution on neural processing unitINTEL CORP·Filed 2025·Application pending·0 cites
- 1662US12229673B2Sparsity-aware datastore for inference processing in deep neural network architecturesINTEL CORP·Filed 2021·Granted Feb 18, 2025·0 cites·25 claims
- 1762US2025371331A1Implementing n:m sparsity in a digital compute-in-memory acceleratorINTEL CORP·Filed 2025·Application pending·0 cites
- 1860US2025307651A1Training and fine-tuning neural network on neural processing unitINTEL CORP·Filed 2025·Application pending·0 cites
- 1958US2024403616A1Approximating activation functions in neural networks with programmable look-up tableINTEL CORP·Filed 2023·Application pending·0 cites
- 2058US2023325665A1Sparsity-based reduction of gate switching in deep neural network acceleratorsINTEL CORP·Filed 2023·Application pending·0 cites
- 2158US2023368030A1Block-wise pruning of weights in deep neural networkRaha Arnab·Filed 2023·Application pending·0 cites
- 2257US2023021396A1Techniques For Increasing Activation Sparsity In Artificial Neural NetworksINTEL CORP·Filed 2022·Application pending·0 cites
- 2357US2023394312A1Pruning activations and weights of neural networks with programmable thresholdsINTEL CORP·Filed 2023·Application pending·0 cites
- 2456US11907827B2Schedule-aware tensor distribution moduleINTEL CORP·Filed 2019·Granted Feb 20, 2024·0 cites·18 claims
- 2555US2023059976A1Deep neural network (dnn) accelerator facilitating quantized inferenceINTEL CORP·Filed 2022·Application pending·0 cites
- 2654US2022188075A1Floating point multiply-accumulate unit for deep learningINTEL CORP·Filed 2022·Application pending·0 cites
- 2754US2024160695A1Approximating activation function in neural network with look-up table having hybrid architectureINTEL CORP·Filed 2023·Application pending·0 cites
- 2854US2023252299A1Detecting and mitigating fault in sparsity computation in deep neural networkKUNDU SHAMIK·Filed 2023·Application pending·0 cites
- 2953US2023140173A1Deep neural network (dnn) accelerators with heterogeneous tilingRaha Arnab·Filed 2022·Application pending·0 cites
- 3053US2022188638A1Data reuse in deep learningINTEL CORP·Filed 2022·Application pending·0 cites
- 3153US2023376274A1Floating-point multiply-accumulate unit facilitating variable data precisionsINTEL CORP·Filed 2023·Application pending·0 cites
- 3253US2024013040A1Output drain path facilitating flexible schedule-based deep neural network acceleratorINTEL CORP·Filed 2023·Application pending·0 cites
- 3353US2024028895A1Switchable one-sided sparsity accelerationRaha Arnab·Filed 2023·Application pending·0 cites
- 3453US2022261623A1System and method for channel-separable operations in deep neural networksINTEL CORP·Filed 2022·Application pending·0 cites
- 3552US12386618B2Multi-buffered register files with shared access circuitsINTEL CORP·Filed 2020·Granted Aug 12, 2025·0 cites·22 claims
- 3652US2022075659A1Runtime configurable register files for artificial intelligence workloadsINTEL CORP·Filed 2021·Application pending·0 cites
- 3751US2023351181A1Approximating activation functions with taylor seriesCHEEMA UMER IFTIKHAR·Filed 2023·Application pending·0 cites
- 3851US2023229917A1Hybrid multipy-accumulation operation with compressed weightsINTEL CORP·Filed 2023·Application pending·0 cites
- 3951US2024119269A1Dynamic sparsity-based acceleration of neural networksRaha Arnab·Filed 2023·Application pending·0 cites
- 4051US2023229507A1Scheduling computations in deep neural network based on sparsityINTEL CORP·Filed 2023·Application pending·0 cites
- 4150US2023017662A1Deep neural network (dnn) accelerators with weight layout rearrangementKADRI SUDHEENDRA·Filed 2022·Application pending·0 cites
- 4250US2023221994A1Dynamic uncompression for channel-separable operation in neural networkRaha Arnab·Filed 2023·Application pending·0 cites
- 4349US2024296961A1System and method for modeling neuronal synaptic functionality on a combined software and hardware architectureUNIV NEW YORK STATE RES FOUND·Filed 2022·Application pending·0 cites
- 4448US2023073661A1Accelerating data load and computation in frontend convolutional layerINTEL CORP·Filed 2022·Application pending·0 cites
- 4547US2023018857A1Sparsity processing on unpacked dataPOWER MARTIN·Filed 2022·Application pending·0 cites
- 4647US2022292366A1Methods and apparatus to perform low overhead sparsity acceleration logic for multi-precision dataflow in deep neural network acceleratorsINTEL CORP·Filed 2022·Application pending·0 cites
- 4747US2024111830A1Accuracy-based approximation of activation functions with programmable look-up table having area budgetINTEL CORP·Filed 2023·Application pending·0 cites
- 4846US10963975B2System architecture and method of processing data thereinPURDUE RESEARCH FOUNDATION·Filed 2019·Granted Mar 30, 2021·0 cites·20 claims
- 4946US2021042617A1Accelerated loading of unstructured sparse data in machine learning architecturesINTEL CORP·Filed 2020·Application pending·0 cites
- 5045US10591902B2Microcontroller energy management systemPURDUE RESEARCH FOUNDATION·Filed 2017·Granted Mar 17, 2020·0 cites·4 claims
Showing the top 50 of 52 patent records by PatentIndex Score.
Join the waitlist — get patent alerts
Get an alert when Arnab Raha 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 →