Interactive quantum computing studio

Quantum Computing Foundations Lab

Explore qubits, superposition, interference, entanglement, gates, measurement, algorithms, and noise in one self-contained teaching workspace. Every simulation runs locally in this HTML file.

Standalone HTML Interactive simulations Student-oriented models
Start here • no prior quantum mechanics required

From one qubit to a working quantum circuit

A quantum computer is not simply a faster conventional computer. It represents and transforms complex probability amplitudes, then uses interference and measurement to extract useful information. This lab builds that idea gradually and lets you test each concept.

Conceptual explanationsState-vector mathematicsCanvas visualizationsStudent exercisesOffline operation
01

Orientation: what quantum computing is

Vocabulary, expectations, and the mathematical language used throughout the page

Quantum computing is a model of computation that stores information in quantum states and manipulates those states with controlled physical operations. A useful quantum algorithm arranges amplitudes so that unwanted answers interfere destructively while useful answers are made more likely.

A quantum processor does not inspect every possible answer and return the whole list. Measurement produces classical outcomes, so an algorithm must be designed to concentrate probability on information that can be recovered with a practical number of measurements.

A productive mental model

Think of a quantum program as an experiment: prepare a state, transform its amplitudes, entangle systems when useful, measure, and repeat enough times to estimate a distribution.

The recurring workflow

classical input → state preparation → gates → interference → measurement → classical output

Most circuit-model tutorials use this sequence. Different hardware platforms implement the physical details differently, but the abstract circuit description is broadly shared.

Learning objectives

|ψ⟩

Represent a qubit

Interpret amplitudes, probabilities, relative phase, normalization, and positions on the Bloch sphere.

H

Explain superposition

Distinguish a coherent superposition from a classical mixture and connect phase to interference.

EPR

Recognize entanglement

Identify Bell states, examine joint outcomes, and understand what correlations do and do not imply.

U

Build circuits

Apply common one- and two-qubit gates and inspect the resulting two-qubit state vector.

M

Analyze measurements

Use the Born rule, estimate sampling uncertainty, and see how finite shots affect observed counts.

QEC

Describe limitations

Relate decoherence, gate errors, connectivity, and error correction to practical quantum computing.

What quantum computers are not

  • They do not make every problem exponentially faster.
  • They do not reveal all branches of a superposition in one measurement.
  • They do not permit faster-than-light messaging through entanglement.
  • They do not eliminate the need for classical computers.

Mathematics you need

Comfort with vectors, complex numbers, trigonometry, probability, and basic matrices is helpful. The simulations provide numerical results so students can begin conceptually and revisit the formalism later.

Minimal mathematical toolkit

IdeaNotationMeaning in this lab
Complex numberz = a + biAmplitudes can have both magnitude and phase. The imaginary unit satisfies i2 = -1.
Magnitude squared|z|2 = a2 + b2Converts a complex amplitude into a nonnegative probability contribution.
Column vector[α, β]TStores amplitudes associated with computational-basis states.
Matrix-vector product|ψ'⟩ = U|ψ⟩Describes how a quantum gate transforms a state.
Tensor product|a⟩ ⊗ |b⟩Combines the state spaces of multiple quantum systems.
Inner product⟨φ|ψ⟩Measures overlap between states and appears in probability calculations.
02

The qubit

A normalized two-dimensional quantum state with amplitude and relative phase

A classical bit is described as 0 or 1. A pure qubit is described by two complex amplitudes associated with basis states |0⟩ and |1⟩. Before measurement, both amplitudes participate in later transformations.

|ψ⟩ = α|0⟩ + β|1⟩,   where   |α|2 + |β|2 = 1

When measured in the computational or Z basis, the probability of 0 is |α|2 and the probability of 1 is |β|2. The amplitudes themselves are not ordinary probabilities because they may be negative or complex and can interfere.

Bloch-sphere form

|ψ⟩ = cos(θ/2)|0⟩ + eiφ sin(θ/2)|1⟩

The polar angle θ controls the balance between 0 and 1. The azimuthal angle φ controls relative phase. A global phase multiplying the entire state cannot be observed directly.

Qubit state controls

Move around the Bloch sphere, inspect amplitudes, and take measurements

PURE STATE
α amplitude0.7071 + 0.0000i
β amplitude0.7071 + 0.0000i
P(0)50.00%
P(1)50.00%
Bloch vector(1.000, 0.000, 0.000)
State label|+⟩

Bloch sphere and measurement

The arrow shows the state; bars show the latest finite-shot experiment

CANVAS
Z basis: north = |0⟩, south = |1⟩
500
0
500
1

Amplitude normalization calculator

Enter any nonzero pair of complex numbers and convert them into a valid qubit

CALCULATOR
Input norm1.414214
Normalized state0.7071|0> + 0.7071|1>
P(0)50.00%
P(1)50.00%

Global phase versus relative phase

The states |ψ⟩ and eiγ|ψ⟩ predict the same measurement statistics in every experiment. By contrast, changing only the phase of β changes the relative phase, which can change interference after another gate.

A qubit is not a tiny classical switch

The Bloch sphere visualizes the abstract state of an ideal two-level quantum system. It is not necessarily the physical path traced by an electron, photon, ion, or superconducting circuit.

03

Superposition and interference

Why phase-sensitive amplitudes are more powerful than ordinary probability weights

A state is in a superposition relative to a chosen basis when it is expressed as a nontrivial linear combination of basis states. For example, |+⟩ = (|0⟩ + |1⟩)/√2 is an equal superposition in the Z basis but is a definite state in the X basis.

The phrase “the qubit is both 0 and 1” can be a useful first metaphor, but it is incomplete. A coherent superposition carries relative phase, and later operations can recombine its amplitudes to produce interference. A classical 50/50 mixture lacks that phase relationship.

The Hadamard gate

H = 1/√2 [ 1  1 ; 1  −1 ]
H|0⟩ = |+⟩    and    H|1⟩ = |−⟩

The Hadamard gate changes between Z-basis and X-basis descriptions. Applying H twice returns the original state because H2 = I.

+

Constructive interference

Amplitudes with aligned phase add, increasing the probability of the associated outcome.

−

Destructive interference

Amplitudes with opposite phase cancel, decreasing or eliminating an outcome.

B

Basis dependence

A state definite in one measurement basis may be a superposition in another.

Single-qubit interference experiment

Simulates the circuit |0⟩ → H → Rz(φ) → H → measure

SIMULATOR
P(0) = [1 + V cos(φ)] / 2,    P(1) = [1 − V cos(φ)] / 2
Ideal P(0)100.00%
Ideal P(1)0.00%
Observed counts0: 1000, 1: 0
InterpretationConstructive at output 0

Interference fringe

The marker follows your phase; lower visibility models loss of coherence

LIVE PLOT
1000
0
0
1

Superposition versus mixture

Both |+⟩ and a 50/50 classical mixture of |0⟩ and |1⟩ produce equal Z-basis outcomes. Apply H before measurement, however, and |+⟩ becomes |0⟩ with certainty while the mixture still gives a 50/50 distribution. Interference reveals coherence.

Why algorithms care about phase

Many quantum algorithms encode information in relative phase and then use a transform to convert phase differences into measurable probability differences. Tracking only probabilities would miss the mechanism.

04

Entanglement

Joint quantum states that cannot be reduced to independent states of their parts

Two qubits are separable when their joint state can be written as a tensor product |a⟩ ⊗ |b⟩. They are entangled when no such factorization exists. In an entangled state, the complete system has a well-defined state even though each qubit by itself may not have a pure state.

The Bell state |Φ+⟩ = (|00⟩ + |11⟩)/√2 is created by applying H to the first qubit and then a CNOT controlled by that qubit. Measuring both in the Z basis produces 00 or 11 with equal probability, never 01 or 10 in the ideal model.

Four Bell states

|Φ+⟩ = (|00⟩ + |11⟩)/√2
|Φ−⟩ = (|00⟩ − |11⟩)/√2
|Ψ+⟩ = (|01⟩ + |10⟩)/√2
|Ψ−⟩ = (|01⟩ − |10⟩)/√2

Correlations are basis-sensitive

Bell states can show perfect correlation or anticorrelation in selected bases. The pattern is stronger than any local hidden-variable description can reproduce across appropriately chosen settings.

No faster-than-light signal

A local measurement outcome is random. Entanglement becomes evident only when classical records from both locations are later compared, so it cannot be used by itself to transmit a controllable message instantaneously.

Bell-state correlation lab

Select a Bell state and local Pauli measurement bases

2 QUBITS
Ideal correlation E+1.000
Observed correlation+1.000
Alice marginal P(+)50.00%
Bob marginal P(+)50.00%

Joint outcome distribution

Outcomes use + and - eigenvalue labels for the selected bases

COUNTS
++
50.0%
+-
0.0%
-+
0.0%
--
50.0%
OutcomeIdeal probabilityObserved count

CHSH inequality calculator

Uses the singlet-state prediction E(a,b) = -cos(a-b) for axes in one plane

BELL TEST
Ideal |S|2.8284
Sampled |S|2.8284
Classical CHSH bound2.0000
Quantum Tsirelson bound2.8284

CHSH result

The classical local-hidden-variable threshold is |S| ≤ 2

CORRELATION
02 classical2.828 quantum4 algebraic
SettingIdeal ESampled E

Ideal violation predicted

These settings produce |S| = 2√2, which exceeds the classical CHSH bound.

05

Quantum gates and circuits

Reversible unitary transformations, controlled operations, and two-qubit state vectors

In the circuit model, a quantum gate is represented by a unitary matrix U satisfying U†U = I. Unitary evolution preserves normalization and is reversible. Measurement is shown separately because it is not an ordinary unitary gate on the measured system.

Single-qubit gates rotate or phase-shift a Bloch vector. Multi-qubit gates condition one operation on another qubit and can create entanglement. A universal gate set can approximate any finite quantum computation to arbitrary accuracy.

State update rule

|ψafter⟩ = U |ψbefore⟩

For two qubits:
|ψ⟩ = a00|00⟩ + a01|01⟩ + a10|10⟩ + a11|11⟩
GateActionSelected relationshipsTypical use
XBit flip; rotation by pi about XX|0> = |1>, X|1> = |0>Prepare or invert a computational-basis value.
YBit and phase flip; rotation about YY|0> = i|1>, Y|1> = -i|0>General rotations and Pauli measurements.
ZPhase flip; rotation about ZZ|+> = |->Change relative phase without changing Z probabilities.
HHadamard basis changeH|0> = |+>, H|+> = |0>Create or recombine equal superpositions.
SQuarter turn about Z, up to global phaseS = diag(1,i)Clifford circuits and phase control.
TEighth turn about Z, up to global phaseT = diag(1,e^(i*pi/4))Non-Clifford resource for universal fault-tolerant circuits.
CNOTFlips target when control is 1CNOT(H|0> tensor |0>) = |Phi+>Entanglement, arithmetic, and error correction.
SWAPExchanges two qubit statesSWAP|01> = |10>Route quantum information across limited connectivity.

Two-qubit circuit workbench

Apply gates from left to right; q0 is the most significant displayed bit

STATE VECTOR

Single-qubit gates on q0

Single-qubit gates on q1

Two-qubit operations

Concurrence0.0000
EntanglementSeparable
Normalization1.000000
Latest samplesNot sampled

Circuit and resulting state

Amplitudes are complex; bars show computational-basis probabilities

LIVE
Initial state |00>
BasisAmplitudePhaseProbability
|00⟩
100.0%
|01⟩
0.0%
|10⟩
0.0%
|11⟩
0.0%

Why concurrence is shown

For a pure two-qubit state with amplitudes a, b, c, and d, concurrence is C = 2|ad - bc|. C = 0 indicates a separable pure state; C = 1 indicates maximal two-qubit entanglement. It does not replace the broader theory of mixed-state entanglement.

No-cloning theorem

There is no unitary operation that perfectly copies an arbitrary unknown quantum state. Known classical basis values can be copied with CNOT, but an arbitrary superposition cannot be cloned without violating linearity.

06

Measurement, sampling, and state space

Born probabilities, finite-shot uncertainty, collapse, tomography, and classical simulation cost

Measurement converts quantum information into a classical outcome. For a state |ψ⟩ measured with projector Pm, the Born rule gives p(m) = ⟨ψ|Pm|ψ⟩. After an ideal projective measurement yielding m, the state is updated consistently with that outcome.

Because individual results are probabilistic, quantum experiments are normally repeated. Each repetition is called a shot. Counts approximate the underlying probabilities, with sampling fluctuations that shrink roughly as 1/√N.

Binomial sampling uncertainty

standard error = sqrt[p(1-p)/N]
approximate 95% margin = 1.96 sqrt[p(1-p)/N]

This simple interval describes ideal independent sampling. Real hardware can add readout bias, drift, correlations, leakage, and other nonideal effects.

Z

Measurement basis

A basis defines which alternatives the instrument distinguishes. Rotating before a Z measurement implements X, Y, or other basis measurements.

N

Repeated shots

One shot gives one bit string. Many shots estimate a probability distribution or expectation value.

ρ

Tomography

Measurements in multiple settings can reconstruct an estimated density matrix, but the number of required settings grows rapidly.

Shot-noise calculator

Estimate statistical uncertainty for a binary measurement

CALCULATOR
Expected successes500.0
Standard error0.015811
Selected margin+/- 0.030990
Approximate interval[0.4690, 0.5310]

Near p = 0 or p = 1, or with very few shots, a normal approximation can be poor. Exact or score-based intervals may be preferable for formal analysis.

State-vector memory calculator

See why exact classical simulation becomes difficult as qubit count grows

2^n AMPLITUDES
Complex amplitudes1,073,741,824
Raw memory16.00 GiB
Basis-state index bits30
Scaling stepEach added qubit doubles memory

Specialized simulators can exploit sparsity, stabilizer structure, tensor networks, symmetry, approximate truncation, and distributed memory. The full dense state-vector estimate is still a useful scaling baseline.

Measurement concepts students often confuse

StatementBetter interpretation
“Measurement merely reads a value that was already present.”For general quantum observables, the theory predicts outcome distributions and post-measurement states; experiments rule out broad classes of simple local preassigned-value models.
“One measurement reveals the amplitudes.”A single shot returns one outcome. Amplitude information requires many preparations and measurements, often in multiple bases.
“More shots remove hardware error.”More shots reduce random sampling uncertainty but do not automatically eliminate systematic bias, decoherence, or calibration error.
“A zero count means zero probability.”With finite sampling, a rare event may simply not appear. Confidence bounds depend on the shot count and model assumptions.
07

Quantum algorithms and speedup

How structure, interference, and measurement can reduce particular computational costs

A quantum advantage comes from an algorithm-problem-hardware combination, not from superposition alone. An algorithm must prepare a useful state efficiently, exploit mathematical structure through coherent operations, and produce an answer that survives measurement with favorable resource scaling.

Complexity claims should specify the task, input model, error tolerance, success probability, and the cost of data loading and readout. Some celebrated speedups require fault-tolerant machines, while near-term approaches often seek useful approximations or scientific simulations.

Three recurring algorithmic ideas

  • Phase kickback: encode function information into phase.
  • Fourier structure: reveal periodicity or eigenphases.
  • Amplitude amplification: rotate probability toward desired states.
DJ

Deutsch-Jozsa

A teaching algorithm that distinguishes constant from balanced oracle functions with one quantum query under a promise, illustrating interference and query complexity.

G

Grover search

Finds a marked item in an unstructured search space using on the order of the square root of N oracle calls, rather than order N classically.

QFT

Phase estimation

Estimates an eigenphase of a unitary operator. It supports Shor's algorithm and many quantum simulation and chemistry methods.

S

Shor factoring

Uses period finding to factor integers in polynomial time in the ideal fault-tolerant circuit model, threatening widely used public-key schemes at sufficient scale.

SIM

Quantum simulation

Models quantum systems using quantum degrees of freedom, a motivation extending from Feynman's early arguments to chemistry and materials research.

VQE

Variational methods

Alternate parameterized quantum circuits with classical optimization. They are flexible, but trainability, noise, benchmarking, and classical competition matter.

Grover amplitude-amplification calculator

Track success probability as the Grover iterate rotates toward marked states

SEARCH MODEL
sin(θ) = sqrt(M/N),    P(k) = sin2[(2k+1)θ]
Search-space size N256
Recommended k12
Success P(k)99.99%
Initial marked probability0.3906%

Grover's algorithm assumes an oracle that identifies marked states coherently. Oracle construction, reversible implementation, and error-correction overhead can dominate a practical application.

Grover success by iteration

Too many iterations rotate amplitude away from the target again

LIVE PLOT

Near the first maximum

The selected iteration count is close to the first high-probability peak.

Comparing selected algorithms

Problem / methodClassical reference scalingQuantum scaling or roleImportant qualification
Unstructured searchO(N)O(sqrt(N)) oracle queriesQuadratic, not exponential; coherent oracle implementation is required.
Integer factoringNo known classical polynomial-time algorithmPolynomial-time quantum algorithm in the standard ideal modelCryptographically relevant instances require large fault-tolerant resources.
Quantum-system dynamicsGeneric state descriptions can grow exponentiallyQuantum hardware naturally represents quantum state spacePreparation, observables, precision, and fault-tolerant cost determine usefulness.
Optimization by QAOA / annealingProblem-dependent heuristicsQuantum heuristic or sampling approachNo blanket speedup; compare solution quality, runtime, energy, and classical baselines.
Linear systems (HHL family)Classical numerical linear algebraPotential exponential dependence improvement under assumptionsReturns a quantum state and relies on sparsity, conditioning, state preparation, and limited observables.
08

Noise, decoherence, and error correction

Why real qubits lose information and how fault-tolerant computation protects logical states

Physical qubits interact imperfectly with controls, neighboring qubits, and their environment. These interactions can change populations, erase relative phase, leak outside the intended two-level subspace, or corrupt measurement. Quantum noise is commonly modeled with channels acting on density matrices.

T1 characterizes energy relaxation toward the ground state in a simple exponential model. T2 characterizes loss of transverse phase coherence. Gate fidelity, crosstalk, connectivity, leakage, readout error, calibration drift, and correlated errors also shape circuit performance.

Simple coherence models

excited population: P1(t) = P1(0)e-t/T1
coherence magnitude: C(t) = C(0)e-t/T2

These exponentials are educational approximations. Actual decay may be nonexponential or depend on pulse sequence, noise spectrum, operating point, and device architecture.

T1

Relaxation

Energy exchange drives an excited state toward a lower-energy state, changing Z-basis populations.

T2

Dephasing

Random phase accumulation reduces coherent off-diagonal terms and weakens interference.

RO

Readout error

The measurement apparatus may report the wrong classical label or disturb neighboring qubits.

XT

Crosstalk

Driving one qubit can unintentionally affect other qubits or couplers.

T1 / T2 coherence explorer

Compare energy relaxation and phase-coherence decay over circuit time

NOISE MODEL
Remaining P10.7788
Remaining coherence0.6997
Relaxed fraction22.12%
Dephased fraction30.03%

Decay curves

Selected time is marked on normalized T1 and T2 exponentials

LIVE PLOT

The condition T2 ≤ 2T1 holds for common Markovian relaxation-plus-dephasing models. Entering T2 above 2T1 is allowed for exploration, but the page will flag that simplified-model inconsistency.

From physical qubits to logical qubits

S

Syndrome extraction

Ancilla-assisted measurements reveal information about errors without directly measuring the encoded logical state.

D

Decoding

A classical decoder interprets syndrome history and selects a likely correction or updates a tracked Pauli frame.

L

Logical operations

Fault-tolerant protocols constrain error propagation while implementing computation on encoded information.

Threshold idea

If physical operations are sufficiently accurate and noise satisfies the code's assumptions, increasing code distance can suppress logical error rates. The required overhead depends strongly on hardware, code, decoder, target algorithm, and error model.

Error mitigation is not error correction

Mitigation techniques estimate or extrapolate less noisy observables without fully encoding a logical qubit. They can help small experiments but usually add sampling or modeling cost and do not by themselves create arbitrarily reliable long computations.

ApproachCore ideaTradeoff
Dynamical decouplingPulse sequences average selected low-frequency noise during idle periods.Adds pulses and depends on timing, calibration, and noise spectrum.
Zero-noise extrapolationRun at several effective noise levels and extrapolate an observable toward zero noise.Sampling overhead and extrapolation bias can grow rapidly.
Probabilistic error cancellationRepresent an ideal operation as a signed combination of noisy operations.Requires noise characterization and can have severe sampling overhead.
Quantum error detectionDiscard runs whose syndrome indicates an error.Reduces accepted data and does not correct every error.
Quantum error correctionEncode logical information across many physical qubits and repeatedly extract syndromes.Large qubit, control, decoding, and fault-tolerant gate overhead.
09

Student activities and notebook

Guided investigations for individual work, lab reports, or classroom discussion

Suggested lab-report structure

State the question, make a prediction, identify variables, record parameter values and results, explain the quantum concept, distinguish ideal-model behavior from hardware behavior, and identify one limitation or follow-on experiment.

Lab 1: Map the Bloch sphere

20-30 min

Question: How do theta and phi separately affect computational-basis measurement?

  1. Test |0>, |1>, |+>, |->, |+i>, and |-i>.
  2. Hold theta at 90 degrees and vary phi.
  3. Explain why P(0) and P(1) stay constant while the state changes.
  4. Apply a basis-changing gate in the circuit lab to expose phase.

Lab 2: Coherence and interference

20-30 min

Question: What information is lost as visibility approaches zero?

  1. Sweep phi with V = 100% and sketch the fringe.
  2. Repeat with V = 50% and V = 0%.
  3. Compare a coherent |+> state with a 50/50 mixture.
  4. Relate fringe contrast to off-diagonal density-matrix terms.

Lab 3: Bell correlations

30-40 min

Question: How can local randomness coexist with strong joint structure?

  1. Measure Phi+ in Z/Z, X/X, Y/Y, and X/Y.
  2. Record joint probabilities and individual marginals.
  3. Repeat for Psi-.
  4. Explain why marginals do not reveal the other party's basis or result.

Lab 4: Build and unbuild entanglement

20-30 min

Question: Which gate sequence creates a Bell pair, and can it be reversed?

  1. Start in |00>, apply H on q0, then CX 0>1.
  2. Verify concurrence 1 and probabilities at 00 and 11.
  3. Apply CX 0>1 again, then H on q0.
  4. Explain why the state returns to |00>.

Lab 5: Finite-shot convergence

20-30 min

Question: How rapidly do observed frequencies approach a model probability?

  1. Choose p = 0.3 in the shot calculator.
  2. Compare 100, 1,000, 10,000, and 100,000 shots.
  3. Run matching samples in a simulator.
  4. Plot absolute frequency error versus the square root of N.

Lab 6: Optimize Grover iterations

25-35 min

Question: Why is there an optimal stopping time?

  1. Use n = 8 with M = 1, 4, and 16.
  2. Compare k - 1, k, and k + 1 around the best integer.
  3. Continue beyond the peak and observe overshoot.
  4. Derive the approximate square-root relationship.

Student notebook

Notes are saved locally in this browser and can be exported as a text file

LOCAL ONLY

No notebook changes saved in this session.

10

Knowledge check

Twelve questions covering the conceptual and mathematical foundations

1. For |psi> = alpha|0> + beta|1>, which condition makes the state normalized?

Probabilities are magnitude squares, so their sum must be one.

2. Which feature distinguishes a coherent equal superposition from a classical 50/50 mixture?

Coherence retains relative phase; a basis change can convert that phase into observable probability differences.

3. What does the Hadamard gate do to |0>?

H maps computational-basis states to equal X-basis superpositions.

4. Which state is entangled?

The Bell state cannot be factored into independent one-qubit states.

5. Why can entanglement not be used alone for faster-than-light communication?

The reduced local statistics do not carry a controllable message about the distant setting or outcome.

6. Which property must an ideal closed-system quantum gate have?

Unitary evolution preserves inner products and normalization and is reversible.

7. Starting from |00>, which sequence creates Phi+?

H creates the branch amplitudes and CNOT correlates the target with the control.

8. Increasing a binary experiment from N to 4N shots changes the standard error by approximately what factor?

Sampling error scales approximately as 1/sqrt(N).

9. A dense n-qubit state vector contains how many complex amplitudes?

Each qubit doubles the computational-basis dimension.

10. What happens if Grover iterations continue far beyond the first optimum?

Amplitude amplification is an oscillatory rotation, so the algorithm must stop near a peak.

11. In a simple exponential model, T1 primarily describes what?

T1 describes decay of excited-state population toward equilibrium in the basic model.

12. Which statement best distinguishes error mitigation from full quantum error correction?

Mitigation can reduce bias for selected observables, but it does not generally provide scalable fault tolerance.
Not gradedAnswer all questions, then grade the knowledge check.
11

Searchable glossary

Core vocabulary used in introductory quantum information science

12

External resources and next study

Official tutorials, browser tools, open learning paths, and algorithm references

Using external resources

The page itself operates offline. The links below open third-party educational sites and may require internet access, accounts, or acceptance of separate terms. Resource availability and interfaces can change over time.

U.S. DOE: Quantum Computing

A concise public-facing explanation of quantum information science and why it matters for scientific computing.

GOVERNMENT EXPLAINER

IBM Quantum Learning

A structured course on the mathematical formulation of quantum information, circuits, entanglement, and protocols.

COURSE PATH

Microsoft Learn: The Qubit

Conceptual and mathematical treatment of qubit states, measurement probabilities, and Bloch-sphere coordinates.

CONCEPT REFERENCE

Google Quantum AI: Cirq

Documentation and tutorials for an open-source Python framework for writing, simulating, and running quantum circuits.

PYTHON FRAMEWORK

Quantum Inspire

A quantum-computing platform with educational materials and access to simulation or supported backends.

PLATFORM

Quantum Country

An interactive essay and spaced-repetition approach to learning quantum computation and quantum mechanics.

INTERACTIVE TEXT

Quantum Algorithm Zoo

A catalog of quantum algorithms organized by application area and computational speedup.

ALGORITHM CATALOG

Suggested next topics

Linear algebra depth

Eigenvalues, spectral decompositions, Kronecker products, density operators, partial trace, and operator norms.

Quantum information

Mixed states, channels, entropy, fidelity, trace distance, teleportation, superdense coding, and no-cloning.

Programming

Build circuits in Qiskit, Cirq, Q#, PennyLane, or another framework; compare ideal and noisy simulation.

Algorithms

Phase estimation, amplitude estimation, quantum walks, Hamiltonian simulation, variational eigensolvers, and cryptanalysis.

Hardware

Superconducting circuits, trapped ions, neutral atoms, photons, spins, topological approaches, controls, and cryogenics.

Fault tolerance

Stabilizer formalism, CSS codes, surface codes, thresholds, lattice surgery, magic-state distillation, and resource estimates.

Selected textbooks and foundational reading

  1. Michael A. Nielsen and Isaac L. Chuang, Quantum Computation and Quantum Information.
  2. Phillip Kaye, Raymond Laflamme, and Michele Mosca, An Introduction to Quantum Computing.
  3. Eleanor G. Rieffel and Wolfgang H. Polak, Quantum Computing: A Gentle Introduction.
  4. Thomas G. Wong, Introduction to Classical and Quantum Computing.
  5. John Preskill, lecture notes on quantum computation and quantum information.
  6. Richard P. Feynman, “Simulating Physics with Computers” (1982).
  7. Peter W. Shor, “Algorithms for Quantum Computation: Discrete Logarithms and Factoring” (1994).
  8. Lov K. Grover, “A Fast Quantum Mechanical Algorithm for Database Search” (1996).