Methods and systems for a synchronized distributed data structure for federated machine learning

Inventors

Stephenson, Mark • Andries, Daina • Aida, Christopher Michio

Assignees

Epidaurus Health Inc

Abstract

A system for an artificial intelligence synchronized distributed ledger. The system includes a computing device containing a receiving module, the receiving module designed and configured to receive an input from a remote device, parse the input to identify protected and non-protected data contained within the input, transform the protected data into a digitally signed assertion and convert the non-protected into an encrypted datastore. The computing device containing a processing module, the processing module designed and configured to receive the digitally signed assertion from the receiving module, insert the digitally signed assertion into an immutable sequential data structure, receive the encrypted datastore, retrieve at least an input, generate a record utilizing the at least a retrieved input, and perform a first machine-learning process utilizing the at least a retrieved input.

Core Innovation

The invention provides a system for a synchronized distributed data structure for federated machine learning in which a computing device receives an input from a remote device and parses the input to identify protected and non-protected data using a language processing module. The language processing module extracts one or more words from the input and produces mathematical associations between the extracted words. The system transforms non-protected data into an encrypted proof-linked assertion and converts protected data into an encrypted datastore.

The system inserts the encrypted proof-linked assertion into a hashed field of an immutable sequential data structure, and retrieves the input from at least one of the immutable sequential data structure and the encrypted datastore. It generates a record as a function of the input and performs a first machine-learning process as a function of the input to determine typical transaction values for a set of participants using training data correlating transaction parameters to transaction values. The system calculates an incentive for each participant type in a given transaction as a function of the typical transaction values and generates a sub-listing of the record as a function of contributions received as a function of the calculated incentive.

The invention further receives a parameter list data input and analyzes the parameter list data input to create fragments including information on a drug, a date the drug was filled at a pharmacy, and information on a treatment the drug was utilized for. It performs a second machine-learning process as a function of the fragments and generates a second record utilizing the second machine-learning process. The disclosure also includes multi-party execution and storing of sub-listings in the immutable sequential data structure, with cryptographic foundations such as elliptic curve cryptography, hash chains, secure proof including zero-knowledge proof, digital signatures, and digital certificates.

Claims Coverage

The document includes two independent claims. Each independent claim centers on synchronized immutable sequential data structure storage with encrypted proof-linked assertions, machine-learning-driven typical transaction values and incentives, and a second machine-learning process based on fragmented parameter-list inputs.

Across the independent claims, the claimed coverage combines language processing that distinguishes protected and non-protected data, encryption and insertion of proof-linked assertions into an immutable sequential data structure, a first machine-learning process to determine typical transaction values and calculate participant incentives that drive sub-listings, and a second machine-learning process operating on fragments of parameter-list data.

Stated Advantages

Documented Applications