Member-only story
QML Series, Part 1: What Quantum Machine Learning Is, and What It Is Not
Topic description
Part 1 of a 10-part series on quantum machine learning (QML), written from a researcher’s point of view, with the hype left out. This part defines what QML actually is, separates it from what it is often claimed to be, and sets up the questions the rest of the series answers: how data gets into a quantum computer, how quantum models are trained, why training often fails, and where quantum advantage in machine learning honestly stands.
Two buzzwords walk into a lab
Quantum computing is hyped. Machine learning is hyped. Put them together and you get a field where the headlines run several years ahead of the results.
That is a shame, because the real field is interesting. It is just interesting for different reasons than the press releases suggest.
Most introductions to QML start with the promise: exponential speedups, smarter models, AI that learns from fewer examples. Then they show a circuit diagram and move on. The reader leaves impressed, and slightly misinformed.
This series starts somewhere else. Before asking what QML can do, it is worth being precise about what it is.
