Naveen Mellempudi

Person

Bangalore, IN

People

Patents Grantslast 30 patents

Information Patent Grant
Incremental precision networks using residual inference and fine-gr...
- Patent number 12,198,055
- Issue date Jan 14, 2025
- Intel Corporation
- Abhisek Kundu
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Instructions to convert from FP16 to BF8
- Patent number 12,135,968
- Issue date Nov 5, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Scaling half-precision floating point tensors for training deep neu...
- Patent number 12,106,210
- Issue date Oct 1, 2024
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Apparatuses, methods, and systems for 8-bit floating-point matrix d...
- Patent number 12,056,489
- Issue date Aug 6, 2024
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Dynamic precision management for integer deep learning primitives
- Patent number 12,033,237
- Issue date Jul 9, 2024
- Intel Corporation
- Naveen K. Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Apparatuses, methods, and systems for 8-bit floating-point matrix d...
- Patent number 12,020,028
- Issue date Jun 25, 2024
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Utilizing structured sparsity in systolic arrays
- Patent number 11,977,885
- Issue date May 7, 2024
- Intel Corporation
- Subramaniam Maiyuran
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Instructions for fused multiply-add operations with variable precis...
- Patent number 11,900,107
- Issue date Feb 13, 2024
- Intel Corporation
- Dipankar Das
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Incremental precision networks using residual inference and fine-gr...
- Patent number 11,893,490
- Issue date Feb 6, 2024
- Intel Corporation
- Abhisek Kundu
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Scaling half-precision floating point tensors for training deep neu...
- Patent number 11,823,034
- Issue date Nov 21, 2023
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Dynamic precision management for integer deep learning primitives
- Patent number 11,669,933
- Issue date Jun 6, 2023
- Intel Corporation
- Naveen K. Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Incremental precision networks using residual inference and fine-gr...
- Patent number 11,556,772
- Issue date Jan 17, 2023
- Intel Corporation
- Abhisek Kundu
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Scaling half-precision floating point tensors for training deep neu...
- Patent number 11,507,815
- Issue date Nov 22, 2022
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Scaling half-precision floating point tensors for training deep neu...
- Patent number 11,501,139
- Issue date Nov 15, 2022
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Conversion hardware mechanism
- Patent number 11,494,163
- Issue date Nov 8, 2022
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Scaling half-precision floating point tensors for training deep neu...
- Patent number 11,468,303
- Issue date Oct 11, 2022
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Optimized compute hardware for machine learning operations
- Patent number 11,334,796
- Issue date May 17, 2022
- Intel Corporation
- Dipankar Das
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Dynamic precision management for integer deep learning primitives
- Patent number 11,321,805
- Issue date May 3, 2022
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Instructions for fused multiply-add operations with variable precis...
- Patent number 11,321,086
- Issue date May 3, 2022
- Intel Corporation
- Dipankar Das
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Technologies for scaling deep learning training
- Patent number 11,068,780
- Issue date Jul 20, 2021
- Intel Corporation
- Naveen K. Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Dynamic precision management for integer deep learning primitives
- Patent number 10,825,127
- Issue date Nov 3, 2020
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Optimized compute hardware for machine learning operations
- Patent number 10,776,699
- Issue date Sep 15, 2020
- Intel Corporation
- Dipankar Das
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Performing power management in a multicore processor
- Patent number 10,775,873
- Issue date Sep 15, 2020
- Intel Corporation
- Victor W. Lee
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Hardware apparatuses and methods relating to elemental register acc...
- Patent number 10,719,317
- Issue date Jul 21, 2020
- Intel Corporation
- Victor Lee
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Dynamic precision management for integer deep learning primitives
- Patent number 10,643,297
- Issue date May 5, 2020
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Instructions for fused multiply-add operations with variable precis...
- Patent number 10,528,346
- Issue date Jan 7, 2020
- Intel Corporation
- Dipankar Das
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Performing power management in a multicore processor
- Patent number 10,234,930
- Issue date Mar 19, 2019
- Intel Corporation
- Victor W. Lee
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Hardware apparatuses and methods relating to elemental register acc...
- Patent number 9,996,347
- Issue date Jun 12, 2018
- Intel Corporation
- Victor Lee
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Grant
Performing power management in a multicore processor
- Patent number 9,910,481
- Issue date Mar 6, 2018
- Intel Corporation
- Victor W. Lee
- G06 - COMPUTING CALCULATING COUNTING

Patents Applicationslast 30 patents

Information Patent Application
DYNAMIC PRECISION MANAGEMENT FOR INTEGER DEEP LEARNING PRIMITIVES
- Publication number 20240412318
- Publication date Dec 12, 2024
- Intel Corporation
- Naveen K. Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
UTILIZING STRUCTURED SPARSITY IN SYSTOLIC ARRAYS
- Publication number 20240320000
- Publication date Sep 26, 2024
- Intel Corporation
- Subramaniam Maiyuran
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
SUPPORTING 8-BIT FLOATING POINT FORMAT OPERANDS IN A COMPUTING ARCH...
- Publication number 20240256274
- Publication date Aug 1, 2024
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
INSTRUCTIONS TO CONVERT FROM FP16 TO BF8
- Publication number 20240248720
- Publication date Jul 25, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
APPARATUSES, METHODS, AND SYSTEMS FOR 8-BIT FLOATING-POINT MATRIX D...
- Publication number 20240241722
- Publication date Jul 18, 2024
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
INCREMENTAL PRECISION NETWORKS USING RESIDUAL INFERENCE AND FINE-GR...
- Publication number 20240160931
- Publication date May 16, 2024
- Intel Corporation
- Abhisek KUNDU
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
INSTRUCTIONS FOR FUSED MULTIPLY-ADD OPERATIONS WITH VARIABLE PRECIS...
- Publication number 20240126544
- Publication date Apr 18, 2024
- Intel Corporation
- Dipankar DAS
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
8-BIT FLOATING POINT SOURCE ARITHMETIC INSTRUCTIONS
- Publication number 20240045654
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
SYSTEMS AND METHODS FOR PERFORMING 8-BIT FLOATING-POINT VECTOR DOT...
- Publication number 20240045689
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
8-BIT FLOATING POINT SCALE AND/OR REDUCE INSTRUCTIONS
- Publication number 20240045682
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
INSTRUCTIONS TO CONVERT FROM FP16 TO FP8
- Publication number 20240045684
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
8-BIT FLOATING POINT FUSED MULTIPLY INSTRUCTIONS
- Publication number 20240045688
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
APPARATUSES, METHODS, AND SYSTEMS FOR 8-BIT FLOATING-POINT MATRIX D...
- Publication number 20240045691
- Publication date Feb 8, 2024
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
INSTRUCTIONS TO CONVERT FROM FP16 TO FP8
- Publication number 20240045677
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
APPARATUSES, METHODS, AND SYSTEMS FOR INSTRUCTIONS FOR STRUCTURED-S...
- Publication number 20240045685
- Publication date Feb 8, 2024
- Intel Corporation
- Menachem Adelman
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
8-BIT FLOATING POINT CLASSIFICATION AND MANIPULATION INSTRUCTIONS
- Publication number 20240045687
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
8-BIT FLOATING POINT SQUARE ROOT AND/OR RECIPROCAL SQUARE ROOT INST...
- Publication number 20240045683
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
8-BIT FLOATING POINT COMPARISON INSTRUCTIONS
- Publication number 20240045681
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
INSTRUCTIONS TO CONVERT FROM FP8
- Publication number 20240045686
- Publication date Feb 8, 2024
- Intel Corporation
- Alexander Heinecke
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
SCALING HALF-PRECISION FLOATING POINT TENSORS FOR TRAINING DEEP NEU...
- Publication number 20230409891
- Publication date Dec 21, 2023
- Intel Corporation
- NAVEEN MELLEMPUDI
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
DYNAMIC PRECISION MANAGEMENT FOR INTEGER DEEP LEARNING PRIMITIVES
- Publication number 20230351542
- Publication date Nov 2, 2023
- Intel Corporation
- Naveen K. Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
APPARATUSES, METHODS, AND SYSTEMS FOR 8-BIT FLOATING-POINT MATRIX D...
- Publication number 20230315450
- Publication date Oct 5, 2023
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
SCALING HALF-PRECISION FLOATING POINT TENSORS FOR TRAINING DEEP NEU...
- Publication number 20230141038
- Publication date May 11, 2023
- Intel Corporation
- NAVEEN MELLEMPUDI
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
INCREMENTAL PRECISION NETWORKS USING RESIDUAL INFERENCE AND FINE-GR...
- Publication number 20230087364
- Publication date Mar 23, 2023
- Intel Corporation
- Abhisek KUNDU
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
SYSTOLIC ARRAY HAVING SUPPORT FOR OUTPUT SPARSITY
- Publication number 20220413803
- Publication date Dec 29, 2022
- Intel Corporation
- Jorge Parra
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
OPTIMIZED COMPUTE HARDWARE FOR MACHINE LEARNING OPERATIONS
- Publication number 20220343174
- Publication date Oct 27, 2022
- Intel Corporation
- Dipankar Das
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
DYNAMIC PRECISION MANAGEMENT FOR INTEGER DEEP LEARNING PRIMITIVES
- Publication number 20220327656
- Publication date Oct 13, 2022
- Intel Corporation
- Naveen K. Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
SUPPORTING 8-BIT FLOATING POINT FORMAT OPERANDS IN A COMPUTING ARCH...
- Publication number 20220318013
- Publication date Oct 6, 2022
- Intel Corporation
- Naveen Mellempudi
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
SCALING HALF-PRECISION FLOATING POINT TENSORS FOR TRAINING DEEP NEU...
- Publication number 20220269931
- Publication date Aug 25, 2022
- Intel Corporation
- NAVEEN MELLEMPUDI
- G06 - COMPUTING CALCULATING COUNTING
Information Patent Application
INSTRUCTIONS FOR FUSED MULTIPLY-ADD OPERATIONS WITH VARIABLE PRECIS...
- Publication number 20220214877
- Publication date Jul 7, 2022
- Intel Corporation
- Dipankar DAS
- G06 - COMPUTING CALCULATING COUNTING

Naveen Mellempudi

Person

Patents Grantslast 30 patents

Incremental precision networks using residual inference and fine-gr...

Instructions to convert from FP16 to BF8

Scaling half-precision floating point tensors for training deep neu...

Apparatuses, methods, and systems for 8-bit floating-point matrix d...

Dynamic precision management for integer deep learning primitives

Apparatuses, methods, and systems for 8-bit floating-point matrix d...

Utilizing structured sparsity in systolic arrays

Instructions for fused multiply-add operations with variable precis...

Incremental precision networks using residual inference and fine-gr...

Scaling half-precision floating point tensors for training deep neu...

Dynamic precision management for integer deep learning primitives

Incremental precision networks using residual inference and fine-gr...

Scaling half-precision floating point tensors for training deep neu...

Scaling half-precision floating point tensors for training deep neu...

Conversion hardware mechanism

Scaling half-precision floating point tensors for training deep neu...

Optimized compute hardware for machine learning operations

Dynamic precision management for integer deep learning primitives

Instructions for fused multiply-add operations with variable precis...

Technologies for scaling deep learning training

Dynamic precision management for integer deep learning primitives

Optimized compute hardware for machine learning operations

Performing power management in a multicore processor

Hardware apparatuses and methods relating to elemental register acc...

Dynamic precision management for integer deep learning primitives

Instructions for fused multiply-add operations with variable precis...

Performing power management in a multicore processor

Hardware apparatuses and methods relating to elemental register acc...

Performing power management in a multicore processor

Patents Applicationslast 30 patents

DYNAMIC PRECISION MANAGEMENT FOR INTEGER DEEP LEARNING PRIMITIVES

UTILIZING STRUCTURED SPARSITY IN SYSTOLIC ARRAYS

SUPPORTING 8-BIT FLOATING POINT FORMAT OPERANDS IN A COMPUTING ARCH...

INSTRUCTIONS TO CONVERT FROM FP16 TO BF8

APPARATUSES, METHODS, AND SYSTEMS FOR 8-BIT FLOATING-POINT MATRIX D...

INCREMENTAL PRECISION NETWORKS USING RESIDUAL INFERENCE AND FINE-GR...

INSTRUCTIONS FOR FUSED MULTIPLY-ADD OPERATIONS WITH VARIABLE PRECIS...

8-BIT FLOATING POINT SOURCE ARITHMETIC INSTRUCTIONS

SYSTEMS AND METHODS FOR PERFORMING 8-BIT FLOATING-POINT VECTOR DOT...

8-BIT FLOATING POINT SCALE AND/OR REDUCE INSTRUCTIONS

INSTRUCTIONS TO CONVERT FROM FP16 TO FP8

8-BIT FLOATING POINT FUSED MULTIPLY INSTRUCTIONS

APPARATUSES, METHODS, AND SYSTEMS FOR 8-BIT FLOATING-POINT MATRIX D...

INSTRUCTIONS TO CONVERT FROM FP16 TO FP8

APPARATUSES, METHODS, AND SYSTEMS FOR INSTRUCTIONS FOR STRUCTURED-S...

8-BIT FLOATING POINT CLASSIFICATION AND MANIPULATION INSTRUCTIONS

8-BIT FLOATING POINT SQUARE ROOT AND/OR RECIPROCAL SQUARE ROOT INST...

8-BIT FLOATING POINT COMPARISON INSTRUCTIONS

INSTRUCTIONS TO CONVERT FROM FP8

SCALING HALF-PRECISION FLOATING POINT TENSORS FOR TRAINING DEEP NEU...

DYNAMIC PRECISION MANAGEMENT FOR INTEGER DEEP LEARNING PRIMITIVES

APPARATUSES, METHODS, AND SYSTEMS FOR 8-BIT FLOATING-POINT MATRIX D...

SCALING HALF-PRECISION FLOATING POINT TENSORS FOR TRAINING DEEP NEU...

INCREMENTAL PRECISION NETWORKS USING RESIDUAL INFERENCE AND FINE-GR...

SYSTOLIC ARRAY HAVING SUPPORT FOR OUTPUT SPARSITY

OPTIMIZED COMPUTE HARDWARE FOR MACHINE LEARNING OPERATIONS

DYNAMIC PRECISION MANAGEMENT FOR INTEGER DEEP LEARNING PRIMITIVES

SUPPORTING 8-BIT FLOATING POINT FORMAT OPERANDS IN A COMPUTING ARCH...

SCALING HALF-PRECISION FLOATING POINT TENSORS FOR TRAINING DEEP NEU...

INSTRUCTIONS FOR FUSED MULTIPLY-ADD OPERATIONS WITH VARIABLE PRECIS...