Simple Algorithm
Quantum Computing · Ground State · 16 min

Qubits & Superposition

The unit of quantum information, and why it is not just a fancy bit.

A classical bit is a light switch: on or off. A qubit is stranger and richer. It is described by two amplitudes, numbers that can be negative or even complex, and that is the seed of everything quantum computers can do.

What you’ll learn

  • Describe a qubit state α∣0⟩+β∣1⟩\alpha\ket{0}+\beta\ket{1} and the normalization rule.
  • Predict measurement statistics from amplitudes using the Born rule.
  • Locate any single-qubit state on the Bloch sphere.
  • Explain why relative phase matters even though a single measurement cannot see it.

Start with a bit

A classical bit has two states, 00 and 11. The only non-trivial thing you can do to one bit is flip it: NOT.

A qubit also has two special states, written ∣0⟩\ket{0} and ∣1⟩\ket{1} (read "ket zero", "ket one"). On the sphere below, ∣0⟩\ket{0} is the north pole and ∣1⟩\ket{1} the south pole. Press X, the quantum NOT gate, and watch the arrow travel.

Superposition: two amplitudes

In general a qubit's state is a superposition

∣ψ⟩=α∣0⟩+β∣1⟩\ket{\psi} = \alpha\ket{0} + \beta\ket{1}

where α\alpha and β\beta are complex numbers called amplitudes. They are not probabilities, but they determine probabilities: the chance of reading 00 is ∣α∣2\abs{\alpha}^2 and of reading 11 is ∣β∣2\abs{\beta}^2. Those must add up to one:

∣α∣2+∣β∣2=1\abs{\alpha}^2 + \abs{\beta}^2 = 1
The normalization condition.
Check your understanding

Which of these is a valid qubit state?

  1. 12∣0⟩+12∣1⟩\tfrac12\ket{0} + \tfrac12\ket{1}
  2. 12∣0⟩−12∣1⟩\tfrac{1}{\sqrt2}\ket{0} - \tfrac{1}{\sqrt2}\ket{1}
  3. 0.6∣0⟩+0.6∣1⟩0.6\ket{0} + 0.6\ket{1}
Show the answer

B. 12∣0⟩−12∣1⟩\tfrac{1}{\sqrt2}\ket{0} - \tfrac{1}{\sqrt2}\ket{1}

∣1/2∣2+∣−1/2∣2=12+12=1\abs{1/\sqrt2}^2 + \abs{-1/\sqrt2}^2 = \tfrac12+\tfrac12 = 1. The minus sign is allowed: amplitudes can be negative. The other two have squared magnitudes summing to 0.50.5 and 0.720.72.

Measurement turns amplitudes into dice

You cannot read α\alpha and β\beta off a qubit. A measurement returns a single bit, 00 or 11, at random, with the probabilities above. This is the Born rule. Afterwards the qubit is in the state you saw: measure again and you get the same answer. That is the collapse.

To learn the probabilities you must prepare many identical qubits and measure each one.

Predict before you look

A qubit is prepared in 0.8 ∣0⟩+0.2 ∣1⟩\sqrt{0.8}\,\ket{0} + \sqrt{0.2}\,\ket{1}. You prepare and measure it 1000 times. About how many 1s do you expect?

Show the answer

200 ones

P(1)=∣0.2∣2=0.2P(1) = \abs{\sqrt{0.2}}^2 = 0.2, so about 0.2×1000=2000.2\times1000 = 200. Each run is random, so you will see roughly 200±13200 \pm 13.

The Bloch sphere

Because only the magnitudes and the relative phase of α,β\alpha,\beta are physical, every single-qubit state can be written with two angles:

∣ψ⟩=cos⁡θ2∣0⟩+eiφsin⁡θ2∣1⟩.\ket{\psi} = \cos\tfrac{\theta}{2}\ket{0} + e^{i\varphi}\sin\tfrac{\theta}{2}\ket{1}.

Treat θ\theta as latitude (from the north pole) and φ\varphi as longitude, and every qubit state becomes a point on a sphere: the Bloch sphere.

StateWhereP(0)P(0) : P(1)P(1)
∣0⟩\ket{0}north pole1 : 0
∣1⟩\ket{1}south pole0 : 1
∣±⟩=12(∣0⟩±∣1⟩)\ket{\pm} = \tfrac{1}{\sqrt2}(\ket{0}\pm\ket{1})equator, ±x\pm x½ : ½
∣±i⟩=12(∣0⟩±i∣1⟩)\ket{\pm i} = \tfrac{1}{\sqrt2}(\ket{0}\pm i\ket{1})equator, ±y\pm y½ : ½

Phase: the hidden dimension

∣+⟩\ket{+} and ∣−⟩\ket{-} give identical 50/50 statistics when measured. Are they really different? Yes! They differ by a relative phase of π\pi (a minus sign), and a gate can reveal it. The Hadamard gate HH maps ∣+⟩→∣0⟩\ket{+}\to\ket{0} and ∣−⟩→∣1⟩\ket{-}\to\ket{1}.

Check your understanding

You are handed a qubit that is either ∣+⟩\ket{+} or ∣−⟩\ket{-}. Which procedure tells you which one with certainty?

  1. Measure it directly in the {∣0⟩,∣1⟩}\{\ket0,\ket1\} basis
  2. Apply HH, then measure
  3. Measure it twice and compare
  4. It is impossible: they are the same state
Show the answer

B. Apply HH, then measure

H∣+⟩=∣0⟩H\ket{+} = \ket{0} and H∣−⟩=∣1⟩H\ket{-} = \ket{1}, so after HH a measurement distinguishes them perfectly. A direct measurement gives 50/50 for both.

Why this is powerful

One qubit needs 2 amplitudes. Two qubits need 4, for ∣00⟩,∣01⟩,∣10⟩,∣11⟩\ket{00},\ket{01},\ket{10},\ket{11}. And nn qubits need 2n2^n. Just 300 qubits need more amplitudes than there are atoms in the observable universe.

But there is a catch: measuring nn qubits yields only nn bits. You cannot simply read out the exponential information. The art of quantum computing is to choreograph amplitudes so that wrong answers cancel and right answers reinforce before you measure. You will meet that choreography, called interference, soon.

Think first If a qubit holds two continuous amplitudes, can you store infinitely many classical bits in one qubit?

No. Holevo's theorem says one qubit can deliver at most one classical bit to whoever measures it. The amplitudes are not directly readable; measurement extracts only a sample.

Key takeaways

  • A qubit state is α∣0⟩+β∣1⟩\alpha\ket{0}+\beta\ket{1} with ∣α∣2+∣β∣2=1\abs{\alpha}^2+\abs{\beta}^2=1.
  • Measurement gives one random bit (Born rule) and collapses the state.
  • Every pure qubit state is a point on the Bloch sphere: θ\theta sets the odds, φ\varphi the relative phase.
  • Relative phase is invisible to a single ZZ measurement but changes how amplitudes interfere.
  • nn qubits carry 2n2^n amplitudes, but measurement reveals only nn bits. Algorithms must use interference.

Check yourself

In the interactive lesson these are scored with your confidence, and what you miss comes back for review.

Question 1

For ∣ψ⟩=12∣0⟩+32∣1⟩\ket{\psi} = \tfrac{1}{2}\ket{0} + \tfrac{\sqrt3}{2}\ket{1}, what is the probability of measuring 11?

  1. 32\tfrac{\sqrt3}{2}
  2. 34\tfrac34
  3. 12\tfrac12
  4. 14\tfrac14
Show the answer

B. 34\tfrac34

P(1)=∣32∣2=34P(1) = \abs{\tfrac{\sqrt3}{2}}^2 = \tfrac34. The amplitude is not the probability; its squared magnitude is.

Question 2

You measure a qubit in state ∣+⟩\ket{+} and get 11. You immediately measure again. What happens?

  1. 11 with certainty
  2. 00 or 11 with probability ½ each
  3. 00 with certainty
  4. The qubit is destroyed
Show the answer

A. 11 with certainty

After the first measurement the state has collapsed to ∣1⟩\ket{1}, so the repeat measurement gives 11 every time.

Question 3

Where on the Bloch sphere is the state 12(∣0⟩+i∣1⟩)\tfrac{1}{\sqrt2}(\ket{0} + i\ket{1})?

  1. North pole
  2. On the equator along +x+x
  3. On the equator along +y+y
  4. South pole
Show the answer

C. On the equator along +y+y

Equal magnitudes put it on the equator (θ=π/2\theta=\pi/2), and the relative phase i=eiπ/2i = e^{i\pi/2} gives longitude φ=π/2\varphi=\pi/2: the +y+y direction, the state ∣+i⟩\ket{+i}.

Question 4

Which pair describes the same physical state?

  1. ∣0⟩+∣1⟩\ket{0}+\ket{1} and ∣0⟩−∣1⟩\ket{0}-\ket{1} (normalized)
  2. ∣1⟩\ket{1} and −∣1⟩-\ket{1}
  3. ∣0⟩\ket{0} and ∣1⟩\ket{1}
  4. ∣+⟩\ket{+} and ∣+i⟩\ket{+i}
Show the answer

B. ∣1⟩\ket{1} and −∣1⟩-\ket{1}

−∣1⟩=eiπ∣1⟩-\ket{1} = e^{i\pi}\ket{1} differs only by a global phase, which no measurement can detect. The others differ by a relative phase or are orthogonal.

Go deeper

Further reading

  • M. A. Nielsen & I. L. Chuang, Quantum Computation and Quantum Information, §1.2 (Cambridge, 10th anniv. ed., 2010).
  • A. Matuschak & M. Nielsen, Quantum Country, an essay with spaced-repetition prompts.