White Plains, NY, United States of America

Marc Phillipe Stoecklin

This inventor holds 2 USPTO granted patents and 1 published patent application. Top assignee: International Business Machines Corporation. Active years: 2021-2023.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Marc Phillipe Stoecklin. Retrieved from https://idiyas.com/inventor/marc-phillipe-stoecklin

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

USPTO Granted Patents = 2 

% Patents Active = 100.0

Average Co-Inventor Count = 3.4

ph-index = 1

Forward Citations = 2(Granted Patents)


Company Filing History:


Years Active: 2021-2023

Loading Chart...
2 patents (USPTO):Explore Patents

Title: Innovations by Marc Phillipe Stoecklin

Introduction

Marc Phillipe Stoecklin is an accomplished inventor based in White Plains, NY (US). He has made significant contributions to the field of technology, particularly in data security and deep learning. With a total of 2 patents, Stoecklin's work focuses on enhancing data confidentiality and integrity.

Latest Patents

Stoecklin's latest patents include "Filesystem view separation for data confidentiality and integrity using lattice-based security domains" and "Protecting deep learning models using watermarking." The first patent describes a stackable filesystem architecture designed to prevent data theft while ensuring file integrity protection. This innovative architecture groups processes into ranked filesystem views, known as 'security domains,' and employs an order theory algorithm to determine the appropriate domain for application execution. The system allows for transparent filesystem operations while maintaining data integrity and confidentiality through layered architecture and view separation.

The second patent provides a framework for verifying the ownership of remotely-deployed deep learning models without compromising model accuracy. This approach involves generating a watermark and embedding it in a local deep neural network (DNN) model. The model is trained to learn both the watermark and a predefined label, which is distinct from the true label of the training data. Black-box verification is then performed by querying the model with a data item containing the watermark to determine if the predefined label is returned.

Career Highlights

Stoecklin is currently employed at International Business Machines Corporation (IBM), where he continues to innovate and contribute to advancements in technology. His work has garnered attention for its practical applications in data security and machine learning.

Collaborations

Some of his notable coworkers include Zhongshu Gu and Heqing Huang, who collaborate with him on various projects within the company.

Conclusion

Marc Phillipe Stoecklin's contributions to technology through his patents reflect his commitment to enhancing data security and deep learning. His innovative approaches are paving the way for future advancements in these critical areas.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…