Skip to main content

Inverse Quantum Fourier Transform

What it means​

The Inverse Quantum Fourier Transform, often written as IQFT or QFT^{-1}, is the operation that undoes the Quantum Fourier Transform.

If QFT is like translating a sentence into another language, IQFT is like translating it back.

This may sound simple, but IQFT is extremely important because many quantum algorithms first move into the Fourier view to process information and then come back to the original view to read the answer.

You might be thinking: if we apply QFT and then undo it, is that not pointless?​

It would be pointless only if nothing important happened in between.

But in real algorithms, the useful work often happens while the state is in the Fourier-style view. That is where hidden phase patterns become easier to manipulate. IQFT is then used to bring that processed information back into a form we can measure.

Why the inverse matters​

In mathematics and computing, a transform is only really useful if we can work with it and, when needed, return to the original form.

That is exactly what IQFT does.

It lets us:

  • undo the QFT,
  • convert phase information into measurable form,
  • and recover useful output states after intermediate processing.

Everyday analogy​

Imagine you compress a sound file into a frequency view so you can study its notes. After you finish the analysis, you still want to play the song normally. You need a way back.

QFT is the move into the frequency-like view. IQFT is the move back.

Circuit structure​

Since IQFT is the inverse of QFT, its circuit is built by:

  • reversing the order of the QFT steps,
  • and replacing each gate by its inverse.

That means:

  • swaps stay swaps,
  • Hadamards stay Hadamards,
  • controlled phase rotations become controlled phase rotations with negative angles.

So if QFT uses a rotation by θ\theta, IQFT uses a rotation by −θ-\theta.

You might be thinking: "Why not just measure after QFT and skip IQFT?" Usually because the answer is still encoded in a form that is hard to read directly. IQFT acts like a decoder that turns that hidden structure into clearer output bits.

Why IQFT appears so often​

Many famous algorithms do not measure directly after creating phase patterns.

Instead they:

  1. create a useful phase structure,
  2. apply IQFT,
  3. measure the output.

IQFT turns the hidden phase pattern into something the measurement can actually reveal.

This is one of the most important mental shifts in quantum computing:

  • the answer may not first appear as a visible bit string,
  • it may first appear as phase,
  • and then another circuit is needed to convert that phase into readable output.

Example of the idea​

Suppose a circuit has encoded some useful information in the phase of a state. If we measured immediately, we might not see that information clearly.

After applying IQFT, the state may become concentrated on basis states that represent the answer more directly.

So IQFT is often less about "doing the hard work" and more about "making the hard work visible."

Relation to phase estimation​

IQFT is one of the key parts of Quantum Phase Estimation.

In that algorithm:

  • controlled unitary operations build a phase pattern,
  • IQFT converts that pattern into a bit string,
  • measurement reads the estimate.

Without IQFT, the hidden phase would remain hard to interpret.

Connection to machine learning​

In machine learning, we sometimes move data into a space where the structure is easier to work with, and then map back to a form that humans or downstream systems can use.

That is a good way to think about IQFT.

It is the "bring it back into readable form" step.

Key idea to remember​

IQFT is the undo operation for QFT, but in practice it is much more than a simple reversal. It is the part that often converts hidden phase structure into an output we can measure and use.