Inventor profile of:

ASHUTOSH GARG

City:

Folsom, California

Country:

United States

Published Applications:

44

Last publication date:

2026-06-25

Top Assignees for applications by ASHUTOSH GARG

The entities that hold a legal rights for patent applications filed by inventor GARG ASHUTOSH:

Recent patent applications by GARG ASHUTOSH

ASHUTOSH GARG from Folsom, US has applied for patents for these inventions. The list has both pending applications and granted patents:

#1 | 2026-06-25
US20260178692A1
Physics

MIXED-PRECISION MATRIX MULTIPLICATION

#2 | 2026-04-02
US20260093488A1
Physics

UTILIZING STRUCTURED SPARSITY IN SYSTOLIC ARRAYS

#3 | 2025-06-26
US20250209564A1
Physics

SPARSE OPTIMIZATIONS FOR A MATRIX ACCELERATOR ARCHITECTURE

#4 | 2025-05-22
US20250166114A1
Physics

ARCHITECTURE FOR BLOCK SPARSE OPERATIONS ON A SYSTOLIC ARRAY

#5 | 2025-04-24
US20250130794A1
Physics

MULTI-FORMAT OPERAND CIRCUIT

#6 | 2025-04-24
US20250130774A1
Physics

FLOATING POINT BIAS SWITCHING

#7 | 2025-04-24
US20250130769A1
Physics

STOCHASTIC ROUNDING CIRCUIT

#8 | 2025-04-24
US20250130767A1
Physics

FLOATING-POINT CONVERSION CIRCUIT

#9 | 2025-03-27
US20250104180A1
Physics

ARCHITECTURE FOR BLOCK SPARSE OPERATIONS ON A SYSTOLIC ARRAY

#10 | 2024-12-26
US20240427847A1
Physics

SCALABLE SPARSE MATRIX MULTIPLY ACCELERATION USING SYSTOLIC ARRAYS WITH FEEDBACK INPUTS

#11 | 2024-10-31
US20240362180A1
Physics

GRAPHICS PROCESSORS AND GRAPHICS PROCESSING UNITS HAVING DOT PRODUCT ACCUMULATE INSTRUCTION FOR HYBRID FLOATING POINT FORMAT

#12 | 2024-09-26
US20240320000A1
Physics

UTILIZING STRUCTURED SPARSITY IN SYSTOLIC ARRAYS

#13 | 2024-05-16
US20240161227A1
Physics

Architecture for block sparse operations on a systolic array

#14 | 2024-02-15
US20240053985A1
Physics

SHARING REGISTER FILE USAGE BETWEEN FUSED PROCESSING RESOURCES

#15 | 2023-11-02
US20230351543A1
Physics

Sparse optimizations for a matrix accelerator architecture

#16 | 2023-09-21
US20230297373A1
Physics

INSTRUCTION AND LOGIC FOR SYSTOLIC DOT PRODUCT WITH ACCUMULATE

#17 | 2023-09-07
US20230281272A1
Physics

Scalable sparse matrix multiply acceleration using systolic arrays with feedback inputs

#18 | 2023-06-22
US20230195685A1
Physics

Graphics processors and graphics processing units having dot product accumulate instruction for hybrid floating point format

#19 | 2022-11-17
US20220365901A1
Physics

Graphics processors and graphics processing units having dot product accumulate instruction for hybrid floating point format

#20 | 2022-10-13
US20220326953A1
Physics

Instructions and logic for vector multiply add with zero skipping

#21 | 2022-08-18
US20220261949A1
Physics

Compiler assisted register file write reduction

#22 | 2022-06-30
US20220206795A1
Physics

SHARING REGISTER FILE USAGE BETWEEN FUSED PROCESSING RESOURCES

#23 | 2022-06-02
US20220171827A1
Physics

Sparse matrix multiplication acceleration mechanism

#24 | 2022-05-19
US20220156343A1
Physics

Scalable sparse matrix multiply acceleration using systolic arrays with feedback inputs

#25 | 2022-04-28
US20220129266A1
Physics

Graphics processors and graphics processing units having dot product accumulate instruction for hybrid floating point format

#26 | 2021-12-02
US20210374897A1
Physics

Sparse optimizations for a matrix accelerator architecture

#27 | 2021-11-11
US20210349966A1
Physics

Scalable sparse matrix multiply acceleration using systolic arrays with feedback inputs

#28 | 2021-10-07
US20210312697A1
Physics

Graphics processors and graphics processing units having dot product accumulate instruction for hybrid floating point format

#29 | 2021-09-30
US20210303299A1
Physics

Instruction and logic for systolic dot product with accumulate

#30 | 2021-06-24
US20210192673A1
Physics

Compiler assisted register file write reduction

#31 | 2021-06-24
US20210191724A1
Physics

Instructions and logic for vector multiply add with zero skipping

#32 | 2021-05-13
US20210141857A1
Physics

Dot product multiplier mechanism

#33 | 2021-04-08
US20210103550A1
Physics

Dot product operations on sparse matrix elements

#34 | 2021-03-25
US20210089301A1
Physics

Sharing register file usage between fused processing resources

#35 | 2021-03-18
US20210081201A1
Physics

Utilizing structured sparsity in systolic arrays

#36 | 2021-03-11
US20210073318A1
Physics

Sparse matrix multiplication acceleration mechanism

#37 | 2021-02-04
US20210035258A1
Physics

Sparse optimizations for a matrix accelerator architecture

#38 | 2020-01-02
US20200004534A1
Physics

Register bank conflict reduction for multi-threaded processor

#39 | 2019-11-28
US20190362460A1
Physics

Software scoreboard information and synchronization

#40 | 2019-10-24
US20190324746A1
Physics

Instruction and logic for systolic dot product with accumulate

#41 | 2019-08-29
US20190265973A1
Physics

Fusion of SIMD Processing Units

#42 | 2019-07-23
US15990328
Physics

Software scoreboard information and synchronization

#43 | 2016-12-29
US20160378432A1
Physics

Handling instructions that require adding results of a plurality of multiplications

#44 | 2016-06-30
US20160189327A1
Physics

Reduced power implementation of computer instructions

InventorID:

1580471 ⎘