Intellectual Property India Publishes Patent Application for 'Zero-Knowledge Proof System For Federated Learning Verification' Filed by Vellore Institute Of Technology

MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123244 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 6, 2025, for 'zero-knowledge proof system for federated learning verification.'

Inventor(s) include P Balakrishnan; A Anny Leema; Akarsh Shukla; Aryan Srivastava; and Aryan Srivastava.

The application for the patent was published on Jan. 2, under issue no. 01/2026.

According to the abstract released by the Intellectual Property India: "The present disclosure provides a zero-knowledge proof system for federated learning verification. The system includes a zero-knowledge succinct non-interactive argument of knowledge (zk-SNARK) circuit configured to encode neural network training operations including forward propagation, backward propagation, and weight updates. The zk-SNARK circuit verifies that final weights result from applying stochastic gradient descent to initial weights using private training data. The circuit receives private inputs including training features and target outputs that remain hidden from verifiers, and public inputs including initial weights, final weights, and network configuration parameters. The system includes a proof generator configured to create cryptographic proofs demonstrating correct execution of stochastic gradient descent on private training data. A proof verifier validates the cryptographic proofs against public inputs without accessing private training data. The system accepts or rejects weight updates based on successful proof verification."

Disclaimer: Curated by HT Syndication.