Missouri Patent of the Month – March 2024
In the realm of data security and privacy, TripleBlind, Inc. introduces a groundbreaking method to safeguard convolutional neural networks (CNNs) while still enabling efficient data processing. This innovative approach revolutionizes how sensitive data can be utilized securely across different entities without compromising confidentiality.
Traditionally, sharing neural networks for collaborative tasks poses significant privacy risks, especially when dealing with proprietary or sensitive information. However, TripleBlind’s method offers a robust solution by dividing the CNN’s filters into distinct sets, ensuring that no single entity can access the complete network architecture or sensitive data.
The process begins with the division of filters in the CNN’s first layer into two sets: a first set and a second set. Each set undergoes separate processing, with the first set applied to the input data, generating a set of outputs. Meanwhile, the second set remains untouched, preserving its integrity and confidentiality.
Crucially, TripleBlind’s method ensures that even in subsequent layers of the neural network, specific weights associated with activated neurons are split into corresponding first and second sets. This ensures that each participating entity retains control over its subset of weights, preventing any single entity from accessing the complete set of neural network parameters.
The method incorporates a secure communication protocol, allowing the exchange of processed outputs between entities without revealing sensitive data or compromising network security. By aggregating outputs from both sets of filters, TripleBlind generates the final output of the CNN, leveraging the combined insights while maintaining data privacy.
TripleBlind’s approach has far-reaching implications across various industries, from healthcare and finance to cybersecurity and research collaborations. By enabling secure collaboration and data sharing, organizations can leverage the power of CNNs without exposing proprietary information or compromising data privacy.
Are you developing new technology for an existing application? Did you know your development work could be eligible for the R&D Tax Credit and you can receive up to 14% back on your expenses? Even if your development isn’t successful your work may still qualify for R&D credits (i.e. you don’t need to have a patent to qualify). To find out more, please contact a Swanson Reed R&D Specialist today or check out our free online eligibility test.
Who We Are:
Swanson Reed is one of the U.S.’ largest Specialist R&D tax advisory firms. We manage all facets of the R&D tax credit program, from claim preparation and audit compliance to claim disputes.
Swanson Reed regularly hosts free webinars and provides free IRS CE and CPE credits for CPAs. For more information please visit us at www.swansonreed.com/webinars or contact your usual Swanson Reed representative.
