International business machines corporation (20240121074). Graph Based Embedding Under Fully Homomorphic Encryption simplified abstract
Contents
- 1 Graph Based Embedding Under Fully Homomorphic Encryption
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 Graph Based Embedding Under Fully Homomorphic Encryption - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Unanswered Questions
- 1.11 Original Abstract Submitted
Graph Based Embedding Under Fully Homomorphic Encryption
Organization Name
international business machines corporation
Inventor(s)
Allon Adir of Kiryat Tivon (IL)
Ramy Masalha of Kafr Qari (IL)
Eyal Kushnir of Kfar Vradim (IL)
Ehud Aharoni of Kfar Saba (IL)
Nir Drucker of Zichron Yaakov (IL)
GUY Moshkowich of Nes Ziyona (IL)
Graph Based Embedding Under Fully Homomorphic Encryption - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240121074 titled 'Graph Based Embedding Under Fully Homomorphic Encryption
Simplified Explanation
The patent application describes mechanisms for fully homomorphic encryption enabled graph embedding. An encrypted graph data structure is received, with encrypted entities and predicates. Sets of entity ciphertexts and predicate ciphertexts are generated based on initial embeddings of entity and predicate features. A machine learning process iteratively updates the embeddings of entity and predicate features to generate a computer model for embedding entities and predicates, resulting in a final embedding output.
- Mechanisms for fully homomorphic encryption enabled graph embedding
- Generation of entity and predicate ciphertexts based on initial embeddings
- Iterative machine learning process to update embeddings of entity and predicate features
- Output of final embedding based on updated embeddings of the computer model
Potential Applications
The technology could be applied in secure data sharing, privacy-preserving machine learning, and encrypted graph analytics.
Problems Solved
The technology addresses the challenge of securely embedding graph data while preserving privacy and confidentiality.
Benefits
The benefits include enhanced data security, privacy protection, and the ability to perform computations on encrypted data without decryption.
Potential Commercial Applications
Potential commercial applications include secure data sharing platforms, encrypted machine learning services, and privacy-focused analytics tools.
Possible Prior Art
One possible prior art is the work on homomorphic encryption and secure graph embedding techniques in the field of cryptography and data security.
Unanswered Questions
How does the technology handle scalability issues in large graph datasets?
The article does not provide details on how the technology addresses scalability challenges when dealing with massive graph data structures.
What are the potential performance implications of using fully homomorphic encryption for graph embedding?
The article does not discuss the potential impact on performance when applying fully homomorphic encryption to graph embedding tasks.
Original Abstract Submitted
mechanisms are provided for fully homomorphic encryption enabled graph embedding. an encrypted graph data structure, having encrypted entities and predicates, is received and, for each encrypted entity, a corresponding set of entity ciphertexts is generated based on an initial embedding of entity features. for each encrypted predicate, a corresponding predicate ciphertext is generated based on an initial embedding of predicate features. a machine learning process is iteratively executed, on the sets of entity ciphertexts and the predicate ciphertexts, to update embeddings of the entity features of the encrypted entities and update embeddings of predicate features of the encrypted predicates, to generate a computer model for embedding entities and predicates. a final embedding is output based on the updated embeddings of the entity features and predicate features of the computer model.
