International business machines corporation (20240313966). TRAINING NEURAL NETWORKS WITH NON-POLYNOMIAL ELEMENTS FOR HOMOMORPHIC ENCRYPTION COMPUTATIONS USING SUB-NETWORKS AND MULTI-LOSS simplified abstract
TRAINING NEURAL NETWORKS WITH NON-POLYNOMIAL ELEMENTS FOR HOMOMORPHIC ENCRYPTION COMPUTATIONS USING SUB-NETWORKS AND MULTI-LOSS
Organization Name
international business machines corporation
Inventor(s)
Itamar Zimerman of Tel-Aviv (IL)
Nir Drucker of Zichron Yaakov (IL)
TRAINING NEURAL NETWORKS WITH NON-POLYNOMIAL ELEMENTS FOR HOMOMORPHIC ENCRYPTION COMPUTATIONS USING SUB-NETWORKS AND MULTI-LOSS - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240313966 titled 'TRAINING NEURAL NETWORKS WITH NON-POLYNOMIAL ELEMENTS FOR HOMOMORPHIC ENCRYPTION COMPUTATIONS USING SUB-NETWORKS AND MULTI-LOSS
The abstract of this patent application describes a system with a processor that can handle non-homomorphic encryption (HE) friendly analytics models containing non-polynomial elements. The processor is designed to train a substitution model by replacing the non-polynomial element with a sub-network featuring a polynomial replacement element.
- Simplified Explanation:
This system involves a processor that can work with encryption-friendly analytics models by replacing non-polynomial elements with polynomial elements.
- Key Features and Innovation:
- Processor capable of handling non-homomorphic encryption-friendly analytics models. - Training a substitution model to replace non-polynomial elements with polynomial elements.
- Potential Applications:
- Secure data analytics - Encrypted machine learning models - Privacy-preserving data processing
- Problems Solved:
- Enhancing the security of analytics models - Enabling secure data processing with encryption
- Benefits:
- Improved data privacy - Enhanced security for sensitive information - Efficient data processing with encryption
- Commercial Applications:
"Secure Data Analytics: Enhancing Privacy and Security in Modern Data Processing"
- Prior Art:
Readers can explore prior research on non-homomorphic encryption and secure data processing techniques.
- Frequently Updated Research:
Stay updated on advancements in encryption-friendly analytics models and secure data processing techniques.
Questions about the technology: 1. How does the processor handle non-polynomial elements in encryption-friendly analytics models? 2. What are the potential implications of using polynomial replacement elements in secure data processing?
Original Abstract Submitted
an example system includes a processor to receive a non-homomorphic encryption (he)-friendly analytics model including a non-polynomial element. the processor is to train a substitution model in which the non-polynomial element of the non-homomorphic encryption (he)-friendly analytics model is replaced with a sub-network including a polynomial replacement element.