Authors:

Yashaswini Subramaniam

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

Sharing sensitive data among AI/ML researchers in nuclear research is vital for collaboration and advancing scientific progress. However, this practice comes with risks, such as reverse engineering and data leaks, which can have severe consequences. This abstract presents a collaborative platform designed to facilitate secure collaboration in the nuclear domain. The sharing of sensitive nuclear data is driven by the need to accelerate research and foster collaboration. However, the potential risks associated with exposing critical nuclear data cannot be overlooked. Unauthorized access, reverse engineering, and data leaks pose significant threats to safety, security, and national interests. The main challenge is to establish a secure collaborative platform that enables researchers to work together without compromising proprietary or confidential information. The research proposes innovative approaches to address this challenge, including synthetic data generation, data masking, and covert embedding techniques. Synthetic data generation allows the creation of realistic but non- sensitive data, enabling researchers to work with representative datasets while ensuring the confidentiality of actual nuclear data. Data masking techniques obscure sensitive information by modifying the data in a way that makes it difficult for unauthorized individuals to decipher or understand the sensitive information, further safeguarding the integrity of shared data. Covert embedding methods embed hidden watermarks or signatures to ensure traceability and authenticity of shared information. By building a collaborative platform that leverages these techniques, the research aims to establish a secure environment for AI/ML researchers in the nuclear domain. The platform ensures the protection of sensitive nuclear data while promoting innovation and facilitating knowledge exchange among researchers.

Source:

Purdue University / 2023

Topics:

No topics listed

Co-authors:

Yashaswini Subramaniam