US 12,333,607 B2
Portfolio value evaluation through quantum polynomial approximation
Gabriele Francesco Maria Agliardi, Milan (IT); Kavitha Hassan Yogaraj, Bangalore (IN); Francesco Tacchino, Rueschlikon (CH); Antonio Mezzacapo, Ardsley, NY (US); Gururaja Hebbar, Frisco, TX (US); Omar Shehab, Yorktown Heights, NY (US); Cortiana Giorgio, Munich (DE); Corey O'Meara, Munich (DE); Kumar Jang Bahadur Ghosh, Essen (DE); and Piergiacomo Sabino, Duesseldorf (DE)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US); and E.ON Digital Technology GmbH, Essen (DE)
Filed by International Business Machines Corporation, Armonk, NY (US); and E.ON Digital Technology GmbH, Essen (DE)
Filed on Apr. 11, 2023, as Appl. No. 18/298,445.
Prior Publication US 2024/0354854 A1, Oct. 24, 2024
Int. Cl. G06Q 40/06 (2012.01); G06N 10/60 (2022.01)
CPC G06Q 40/06 (2013.01) [G06N 10/60 (2022.01)] 17 Claims
OG exemplary drawing
 
1. A system, comprising:
a memory that stores computer executable components; and
a processor, operably coupled to the memory, and that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
an evaluating component that employs a quantum register to output a processed vector based on a first variable, of a set of variables indexed over a selected dimension;
a valuation component that approximates a value function and that, based on the approximating, outputs a valuation amount representing provision of an aspect over the selected dimension,
wherein the value function is a function of an inner product of a set of processing functions that are based on the set of variables, including the first variable,
wherein a first processing function of the set of processing functions and of the inner product is based on the processed vector, and
wherein the set of variables correspond to conditions defining the provision of the aspect; and
an encoding component that generates a quantum circuit that encodes a dataset into the quantum register in the form of one or more corresponding quantum states,
wherein the encoding comprises employing a set of initial vectors representing the dataset, indexed over the selected dimension, and having been normalized.