Unsupervised Clustering via Quantum State Entanglement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current unsupervised clustering methods in hybrid computing systems, using a classical digital computer and a quantum computer, are limited by the number of qubits available, restricting the number of data points that can be clustered due to the limitation on the number of qubits a quantum computer can have.
Innovation Solution
The method involves a classical digital computer importing data points, locating them on a Bloch sphere, defining a cost function, and using a quantum computer with a quantum circuit to optimize clustering by entangling qubits, allowing for the processing of multiple data points with fewer qubits through optimized variational parameters and layers in the quantum circuit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a qubit in the quantum computer is used to represent each of the data points in the data set, then unsupervised clustering can be performed, but the number of data points that can be clustered is limited by the number of qubits available
Solution Approach 1:
The quantum computer performs multiple functions: it represents data points as quantum states, executes clustering algorithms through quantum circuits, and processes multiple data points sequentially. This multi-functionality allows the system to overcome the limitation where one qubit equals one data point, enabling the quantum computer to handle more data points than the number of physical qubits by reusing qubits across different data points through quantum state manipulation
Solution Approach 2:
The patent introduces a temporal dimension by processing data points sequentially over multiple time steps. Instead of requiring simultaneous representation of all data points (spatial dimension), the system processes data points one after another, using the same qubits repeatedly. This transforms the problem from a spatial resource constraint to a temporal process, effectively increasing the number of data points that can be clustered beyond the number of available qubits
2Productivity
If the number of qubits is increased to cluster more data points, then the clustering capacity increases, but the device complexity and resource requirements increase
Solution Approach 1:
The quantum computer leverages its own quantum mechanical properties (superposition, entanglement, interference) to perform clustering computations that would require many more classical resources. By using quantum algorithms that exploit these inherent properties, the system achieves high clustering capacity without proportionally increasing the number of physical qubits, as the quantum operations themselves provide the computational power needed
Data Source
AI summary
Method for performing unsupervised clustering of a data set including a plurality of data points by means of minimization of a cost function by a classical digital computer. The classical digital computer locates each data point on a Bloch sphere and sends the position of the data point to a quantum computer. The quantum computer translates each position to a quantum state of a plurality of qubits and implements a quantum circuit on them. The quantum circuit modifies the quantum state of the qubits to a final quantum state based on a plurality of optimized variational parameters provided by an optimizer of the cost function performed in the classical digital computer, the final quantum state corresponding to the label in which each data point is clustered.