MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202621100333 A) filed by Indian Institute Of Technology Gandhinagar on August 19, 2026, for Method, Apparatus And System For Generating Synthetic Multivariate Time-Series Data Using Quantum Circuits.
Inventors include Kumar, Shashi; Sharma, Yash; and Kulkarni, Sameer Gundurao.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: ABSTRACT METHOD, APPARATUS AND SYSTEM FOR GENERATING SYNTHETIC MULTIVARIATE TIME-SERIES DATA USING QUANTUM CIRCUITS A computer-implemented method (100), system, and apparatus for generating synthetic multivariate time-series data via a quantum processor (504), comprising: quantile-binning (102), by a classical processor (502), values of channels at time steps to obtain quantile indices; encoding (104), by the classical processor (502), the quantile indices using binary-reflected Gray code to obtain binary strings, wherein adjacent quantile indices differ by one bit; computing (106), by the classical processor (502), a weighting matrix as an inverse empirical covariance matrix of residuals of weight-one and weight-two Pauli-Z correlators of the binary strings; training (108), by the classical processor (502), a mixture of instantaneous quantum polynomial-time circuits by minimizing a weighted quadratic loss between model correlators obtained through classical Fourier Monte Carlo sampling and empirical correlators; and compiling (110), by the classical processor (502), trained parameters into a deferred-measurement quantum circuit for execution by the quantum processor (504) to generate synthetic time-series samples.
Disclaimer: Curated by HT Syndication.
