Pritish Narayanan

This inventor holds 29 USPTO granted patents and 7 published patent applications, plus 1 CIPO patent and 2 EPO patents, primarily in Neuromorphic Computing. Top assignees: International Business Machines Corporation, Polytechnic University of Turin. Active years: 2019-2026.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Pritish Narayanan. Retrieved from https://idiyas.com/inventor/pritish-narayanan

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

USPTO Granted Patents = 29 (Patent Applications = 7)

% Patents Active = 96.6

CIPO Patents = 1

EPO Patents = 2 (Patent Applications = 4)

Average Co-Inventor Count = 2.8

ph-index = 2

Forward Citations = 12(Granted Patents)

Company Filing History:

29 out of 166,325

International Business Machines Corporation

patents

1 out of 60

Polytechnic University of Turin

patents

where one patent can have more than one assignee

Years Active: 2019-2026

Loading Chart...

Areas of Expertise:

Neuromorphic Arrays

Analog Artificial Intelligence

Deep Learning Operations

Compute-in-Memory

Convolutional Neural Networks

Pulse Width Modulation

Three-Dimensional Mesh

Activation Functions

Weight Matrices Compression

Resistive Processing Units

Signal Amplitude Control

Non-Volatile Memory

29 patents (USPTO):Sort by Forward CitationsExplore Patents

1. 12626111 - Calibrating peripheral variability (2026-05-12),

2. 12596921 - Stochastic bitstream generation with in-situ function mapping (2026-04-07),

3. 12547884 - Hardware implementation of activation functions (2026-02-10),

4. 12518149 - Implicit vector concatenation within 2D mesh routing (2026-01-06),

5. 12456043 - Two-dimensional mesh for compute-in-memory accelerator architecture (2025-10-28),

6. 12255656 - Split pulse width modulation to reduce crossbar array integration time (2025-03-18),

7. 12050997 - Row-by-row convolutional neural network mapping for analog artificial intelligence network training (2024-07-30),

8. 12019590 - System, method and article of manufacture for synchronization-free transmittal of neuron values in a hardware artificial neural networks (2024-06-25),

9. 12003240 - Analog memory-based complex multiply-accumulate (MACC) compute engine (2024-06-04),

10. 11977974 - Compression of fully connected / recurrent layers of deep network(s) through enforcing spatial locality to weight matrices and effecting frequency compression (2024-05-07),

11. 11868893 - Efficient tile mapping for row-by-row convolutional neural network mapping for analog artificial intelligence network inference (2024-01-09),

12. 11823740 - Selective application of multiple pulse durations to crossbar arrays (2023-11-21),

13. 11797833 - Competitive machine learning accuracy on neuromorphic arrays with non-ideal non-volatile memory devices (2023-10-24),

14. 11646944 - Configuring computing nodes in a three-dimensional mesh topology (2023-05-09),

15. 11580373 - System, method and article of manufacture for synchronization-free transmittal of neuron values in a hardware artificial neural networks (2023-02-14),

Title: Innovations and Contributions of Pritish Narayanan in Artificial Intelligence

Introduction

Pritish Narayanan is a prominent inventor located in San Jose, CA, known for his significant contributions to the field of artificial intelligence and neural networks. With an impressive portfolio of 23 patents, Narayanan has made substantial advancements that enhance computational efficiency in artificial intelligence applications.

Latest Patents

Narayanan’s latest innovations include several noteworthy patents. One such patent is titled "Row-by-row convolutional neural network mapping for analog artificial intelligence network training." This patent outlines a computer-implemented method for executing a convolutional neural network (CNN) using a crosspoint array. It describes the configuration of the array to implement convolution layers by storing weights in crosspoint devices, training the CNN, and employing a row-by-row approach to mapping input data.

Another notable patent is for a "System, method and article of manufacture for synchronization-free transmittal of neuron values in hardware artificial neural networks." This invention facilitates computations within artificial neural networks using neurons and synapses, effectively transferring neuron values from one array to another without global clock synchronization. The technology promotes better efficiency and accuracy in neural network operations.

Career Highlights

Pritish Narayanan has had a distinguished career, having worked with renowned institutions such as IBM and Polytechnic University of Turin. His experiences in these environments have greatly influenced his research and development endeavors in the field of artificial intelligence.

Collaborations

Throughout his career, Narayanan has collaborated with esteemed peers, including Geoffrey W. Burr and Hsin-Yu Tsai. These collaborations have contributed to his innovative research and the development of several groundbreaking technologies in the realm of neural networks and artificial intelligence.

Conclusion

With a blend of innovative thinking and extensive expertise, Pritish Narayanan has significantly impacted the artificial intelligence landscape. His latest patents exemplify a commitment to advancing technology while collaborating with other prominent experts in the field, paving the way for future advancements in AI and computational methods.

Profile summary based on public USPTO records.