Inventor profile of:

Dipankar Das

City:

Pune

Country:

India

Published Applications:

65

Last publication date:

2026-04-30

Top Assignees for applications by Dipankar Das

The entities that hold a legal rights for patent applications filed by inventor Das Dipankar:

Recent patent applications by Das Dipankar

Dipankar Das from Pune, IN has applied for patents for these inventions. The list has both pending applications and granted patents:

#1 | 2026-04-30
US20260119902A1
Physics

COMMUNICATION OPTIMIZATIONS FOR DISTRIBUTED MACHINE LEARNING

#2 | 2026-04-02
US20260093488A1
Physics

UTILIZING STRUCTURED SPARSITY IN SYSTOLIC ARRAYS

#3 | 2026-01-08
US20260010969A1
Physics

DYNAMIC PRECISION MANAGEMENT FOR INTEGER DEEP LEARNING PRIMITIVES

#4 | 2026-01-08
US20260010345A1
Physics

SYSTOLIC ARRAY HAVING SUPPORT FOR OUTPUT SPARSITY

#5 | 2026-01-01
US20260004383A1
Physics

ABSTRACTION LAYERS FOR SCALABLE DISTRIBUTED MACHINE LEARNING

#6 | 2025-11-27
US20250363355A1
Physics

HARDWARE IMPLEMENTED POINT TO POINT COMMUNICATION PRIMITIVES FOR MACHINE LEARNING

#7 | 2025-11-27
US20250362924A1
Physics

OPTIMIZED COMPUTE HARDWARE FOR MACHINE LEARNING OPERATIONS

#8 | 2025-07-03
US20250217142A1
Physics

INSTRUCTIONS FOR FUSED MULTIPLY-ADD OPERATIONS WITH VARIABLE PRECISION INPUT OPERANDS

#9 | 2025-06-19
US20250200696A1
Physics

DATA PARALLELISM AND HALO EXCHANGE FOR DISTRIBUTED MACHINE LEARNING

#10 | 2025-05-29
US20250173567A1
Physics

INCREMENTAL PRECISION NETWORKS USING RESIDUAL INFERENCE AND FINE-GRAIN QUANTIZATION

#11 | 2025-02-20
US20250061318A1
Physics

SCALING HALF-PRECISION FLOATING POINT TENSORS FOR TRAINING DEEP NEURAL NETWORKS

#12 | 2024-12-12
US20240412318A1
Physics

DYNAMIC PRECISION MANAGEMENT FOR INTEGER DEEP LEARNING PRIMITIVES

#13 | 2024-09-26
US20240320000A1
Physics

UTILIZING STRUCTURED SPARSITY IN SYSTOLIC ARRAYS

#14 | 2024-05-16
US20240160931A1
Physics

Incremental precision networks using residual inference and fine-grain quantization

#15 | 2024-04-18
US20240126544A1
Physics

Instructions for fused multiply-add operations with variable precision input operands

#16 | 2024-04-11
US20240118892A1
Physics

APPARATUSES, METHODS, AND SYSTEMS FOR NEURAL NETWORKS

#17 | 2024-02-29
US20240070799A1
Physics

ABSTRACTION LAYERS FOR SCALABLE DISTRIBUTED MACHINE LEARNING

#18 | 2023-12-21
US20230409891A1
Physics

Scaling half-precision floating point tensors for training deep neural networks

#19 | 2023-11-23
US20230376762A1
Physics

COMMUNICATION OPTIMIZATIONS FOR DISTRIBUTED MACHINE LEARNING

#20 | 2023-11-02
US20230351542A1
Physics

Dynamic precision management for integer deep learning primitives

#21 | 2023-06-08
US20230177328A1
Physics

HARDWARE IMPLEMENTED POINT TO POINT COMMUNICATION PRIMITIVES FOR MACHINE LEARNING

#22 | 2023-05-11
US20230141038A1
Physics

Scaling half-precision floating point tensors for training deep neural networks

#23 | 2023-03-23
US20230087364A1
Physics

Incremental precision networks using residual inference and fine-grain quantization

#24 | 2022-12-29
US20220413803A1
Physics

SYSTOLIC ARRAY HAVING SUPPORT FOR OUTPUT SPARSITY

#25 | 2022-11-17
US20220366526A1
Physics

Data parallelism and halo exchange for distributed machine learning

#26 | 2022-10-27
US20220343174A1
Physics

Optimized compute hardware for machine learning operations

#27 | 2022-10-13
US20220327656A1
Physics

Dynamic precision management for integer deep learning primitives

#28 | 2022-10-13
US20220326953A1
Physics

Instructions and logic for vector multiply add with zero skipping

#29 | 2022-08-25
US20220269931A1
Physics

Scaling half-precision floating point tensors for training deep neural networks

#30 | 2022-08-04
US20220245454A1
Physics

Communication optimizations for distributed machine learning

#31 | 2022-07-07
US20220214877A1
Physics

Instructions for fused multiply-add operations with variable precision input operands

#32 | 2022-04-14
US20220115362A1
Electricity

SCALABLE HIGH-PERFORMANCE PACKAGE ARCHITECTURE USING PROCESSOR-MEMORY-PHOTONICS MODULES

#33 | 2022-03-31
US20220101480A1
Physics

Abstraction layers for scalable distributed machine learning

#34 | 2022-02-17
US20220050683A1
Physics

APPARATUSES, METHODS, AND SYSTEMS FOR NEURAL NETWORKS

#35 | 2021-12-09
US20210382719A1
Physics

Apparatuses, methods, and systems for access synchronization in a shared memory

#36 | 2021-11-11
US20210350212A1
Physics

ABSTRACTION LIBRARY TO ENABLE SCALABLE DISTRIBUTED MACHINE LEARNING

#37 | 2021-11-04
US20210342692A1
Physics

TECHNOLOGIES FOR SCALING DEEP LEARNING TRAINING

#38 | 2021-06-24
US20210191724A1
Physics

Instructions and logic for vector multiply add with zero skipping

#39 | 2021-04-15
US20210110508A1
Physics

Dynamic precision management for integer deep learning primitives

#40 | 2021-04-15
US20210109888A1
Physics

PARALLEL PROCESSING BASED ON INJECTION NODE BANDWIDTH

#41 | 2021-03-18
US20210081201A1
Physics

Utilizing structured sparsity in systolic arrays

#42 | 2021-03-11
US20210072955A1
Physics

Conversion hardware mechanism

#43 | 2021-01-21
US20210019631A1
Physics

Optimized compute hardware for machine learning operations

#44 | 2020-08-20
US20200265545A1
Physics

Dynamic precision management for integer deep learning primitives

#45 | 2020-08-13
US20200257527A1
Physics

Instructions for fused multiply-add operations with variable precision input operands

#46 | 2019-11-21
US20190354846A1
Physics

Scaling half-precision floating point tensors for training deep neural networks

#47 | 2019-10-03
US20190303743A1
Physics

APPARATUSES, METHODS, AND SYSTEMS FOR NEURAL NETWORKS

#48 | 2019-08-08
US20190243651A1
Physics

Apparatuses, methods, and systems for access synchronization in a shared memory

#49 | 2019-07-25
US20190227797A1
Physics

Apparatus and method for vector multiply and accumulate of packed words

#50 | 2019-07-04
US20190205745A1
Physics

Communication optimizations for distributed machine learning

#51 | 2019-02-07
US20190042939A1
Physics

Circuitry for low-precision deep learning

#52 | 2019-02-07
US20190042242A1
Physics

Instructions for fused multiply-add operations with variable precision input operands

#53 | 2019-02-07
US20190042236A1
Physics

Apparatus and method for vector multiply and accumulate of packed bytes

#54 | 2018-11-08
US20180322607A1
Physics

Dynamic precision management for integer deep learning primitives

#55 | 2018-11-08
US20180322606A1
Physics

Data parallelism and halo exchange for distributed machine learning

#56 | 2018-11-08
US20180322390A1
Physics

Optimized compute hardware for machine learning operations

#57 | 2018-11-08
US20180322387A1
Physics

Hardware implemented point to point communication primitives for machine learning

#58 | 2018-11-08
US20180322382A1
Physics

Scaling half-precision floating point tensors for training deep neural networks

#59 | 2018-11-01
US20180314940A1
Physics

Incremental precision networks using residual inference and fine-grain quantization

#60 | 2018-10-11
US20180293493A1
Physics

Abstraction layers for scalable distributed machine learning

#61 | 2018-10-11
US20180293492A1
Physics

Abstraction library to enable scalable distributed machine learning

#62 | 2018-10-04
US20180285733A1
Physics

Technologies for scaling deep learning training

#63 | 2014-01-16
US20140019668A1
Physics

Memory management systems and methods for embedded systems

#64 | 2013-05-16
US20130124917A1
Physics

Recovering from stack corruption faults in embedded software systems

#65 | 2011-10-06
US20110246831A1
Physics

Method and apparatus for operational-level functional and degradation fault analysis

InventorID:

2314522 ⎘