Resources:
- “Classical ML review | IBM Quantum Learning,” IBM Quantum Learning, 2018. https://quantum.cloud.ibm.com/learning/en/courses/quantum-machine-learning/classical-ml-review (accessed Mar. 04, 2026).
The Concept: Quantum Feature Maps
In classical machine learning, a “kernel trick” is often used to map data into a higher-dimensional space. Quantum computers provide a natural way to do this:
Feature Map
The process of encoding a classical input vector \( \vec{x} \) into a quantum state \( |\Phi(\vec{x})\rangle \) acts as a feature map.
High Dimensionality
Because Hilbert space grows exponentially with the number of qubits, quantum states represent data in a massive space that would be computationally impossible for classical computers to handle.