Each card keeps the plain sentence and the technical definition.
Classical bit
Plain
A classical bit is the smallest unit of ordinary computer information. After you look at it, it is either 0 or 1.
Technical
A bit is a binary variable with a definite value in {0, 1}. Classical circuits transform bits with logic gates such as AND, OR, and NOT.
Why it matters
Almost every comparison in this workshop starts here. Quantum computing adds a different kind of information. It does not throw the bit away.
Common misconception
A qubit is a bit that can be both numbers in a way that tries every answer at once. That picture is easy to pick up, and we will build a clearer one together.
Qubit
KYOO-bit
Plain
A qubit is the basic unit of quantum information. Before measurement it is described by a quantum state, not by a single stored 0 or 1.
Technical
A qubit’s pure state is a normalized vector in a two-dimensional complex vector space, spanned by the basis states |0⟩ and |1⟩.
Why it matters
Every circuit in Composer and Qiskit changes qubit states and then measures them.
Common misconception
A qubit is a tiny classical bit that spins until you look. The useful description is a statevector, not a spinning coin.
State
Plain
The state is the complete description of what we know about the qubits before we measure them.
Technical
For an isolated pure state of n qubits, the state is a normalized vector with 2ⁿ amplitudes, one for each computational basis string.
Why it matters
Gates change the state. Measurement samples it. If you skip the state, the histogram looks like magic.
Common misconception
The state is the histogram. The histogram is a record of many measurements. The state is the description before those measurements.
Statevector
STATE-vek-ter
Plain
A statevector is the list of amplitudes that describes a quantum state.
Technical
In the computational basis, an n-qubit statevector is a column of 2ⁿ complex numbers whose squared magnitudes sum to 1.
Why it matters
In the Python lab we print the statevector before we measure, so you can see the state Composer is also tracking.
Common misconception
A longer statevector means a better quantum computer. The length is 2ⁿ because there are 2ⁿ basis states, not because the device is more powerful by itself.
Basis state
BAY-sis
Plain
A basis state is one of the definite labels we use to write a state, such as |0⟩, |1⟩, |00⟩, or |11⟩.
Technical
The computational basis is the orthonormal set of states labeled by classical bit strings. Any statevector is a weighted sum of these states.
Why it matters
Gates and measurements are defined by what they do to basis states.
Common misconception
Only basis states are real. Superpositions are also valid states. A basis state is a convenient label.
Amplitude
Plain
An amplitude is the complex number in front of a basis state. It is not the probability itself.
Technical
If the state is Σ αₓ |x⟩, the amplitude αₓ is complex. The probability of reading x is |αₓ|², provided the state is normalized.
Why it matters
Two states can have the same probabilities and different amplitudes. Interference depends on that difference.
Common misconception
Amplitude and probability are the same number. You square the magnitude of the amplitude to get the probability.
Probability
Plain
A probability is the chance of a particular measurement result. It is between 0 and 1, and all the chances together add up to 1.
Technical
For a computational-basis measurement, P(x) = |αₓ|². In an ideal simulation with many shots, the fraction of shots equal to x approaches P(x).
Why it matters
Histograms are pictures of probabilities estimated from shots, not the state itself.
Common misconception
If the probability is one half, a single shot must come out half 0 and half 1. One shot is one result.
Phase
fayz
Plain
Phase is the part of an amplitude that changes how states combine, even when the probabilities look the same.
Technical
A complex amplitude can be written as r·e^{iφ}. The angle φ is the phase. Relative phases between amplitudes control interference.
Why it matters
Some gates change probabilities. Some mainly change phase. Both matter.
Common misconception
Phase is a physical angle you can see on the chip. It is part of the mathematical description of the state.
Superposition
SOO-per-puh-ZISH-un
Plain
A superposition is a state that is a combination of basis states, with amplitudes, before measurement.
Technical
A state is a superposition when more than one computational-basis amplitude is nonzero. Measurement in that basis yields one basis state, with probability given by the squared magnitudes.
Why it matters
The Hadamard gate in the Bell lab creates a superposition on qubit 0. Without it, the CNOT has nothing interesting to correlate.
Common misconception
Superposition means the qubit is 0 and 1 at the same time and therefore tries every answer. It means the state is a weighted combination, and a measurement returns one outcome.
Entanglement
en-TANG-gull-ment
Plain
Entanglement is a correlation between qubits that cannot be written as each qubit having its own separate state.
Technical
A multi-qubit pure state is entangled when it cannot be factored into a tensor product of single-qubit states. The Bell state (|00⟩ + |11⟩)/√2 is entangled.
Why it matters
The Composer lab builds this correlation with H and CNOT. The histogram is a clue, not a full proof.
Common misconception
Seeing only 00 and 11 once proves entanglement forever. Those counts are consistent with this preparation. A single computational-basis histogram is not a complete experimental proof.
Quantum gate
Plain
A quantum gate is an operation that changes the state of one or more qubits.
Technical
A gate is a unitary operator on the state space. It is reversible, unlike a measurement.
Why it matters
Circuits are ordered lists of gates, then optional measurements.
Common misconception
A gate measures the qubit. A gate changes the state. Measurement is a separate step.
Hadamard gate
HAD-uh-mard
Plain
The Hadamard gate, written H, takes |0⟩ into an equal superposition of |0⟩ and |1⟩.
Technical
H|0⟩ = (|0⟩ + |1⟩)/√2 and H|1⟩ = (|0⟩ − |1⟩)/√2. Applying H twice returns the original basis state.
Why it matters
It is the first gate in the Bell-state lab.
Common misconception
H always randomizes a qubit. On |0⟩ it creates equal probabilities. On a superposition it can also undo that superposition.
X gate
Plain
The X gate flips |0⟩ and |1⟩, like a classical NOT, when the qubit is in a basis state.
Technical
X is the Pauli X operator. X|0⟩ = |1⟩ and X|1⟩ = |0⟩. On a superposition it exchanges the amplitudes of |0⟩ and |1⟩.
Why it matters
One of the beginner challenges replaces H with X so you can see the Bell lab fall apart in a predictable way.
Common misconception
X creates superposition. X flips basis states. H creates the equal superposition we use here.
CNOT or CX gate
SEE-not, or controlled-X
Plain
CNOT, also called CX, flips the target qubit when the control qubit is |1⟩.
Technical
CX|00⟩ = |00⟩, CX|01⟩ = |01⟩, CX|10⟩ = |11⟩, CX|11⟩ = |10⟩, with the control written first in this sentence. In Qiskit’s little-endian bitstrings, read the rightmost bit as qubit 0.
Why it matters
H on qubit 0 followed by CX onto qubit 1 prepares the Bell state in this lab.
Common misconception
CNOT copies a qubit. It correlates the target with the control. Quantum information cannot be cloned.
Control qubit
Plain
The control qubit is the qubit that decides whether a controlled gate acts.
Technical
In CX, the control selects the subspace in which X is applied to the target. The control itself is not flipped by an ideal CX.
Why it matters
In the lab, qubit 0 is the control. Swapping control and target prepares a different wiring, even if the picture looks similar.
Common misconception
The control is always the top wire. It is whichever qubit you attach the solid dot to.
Target qubit
Plain
The target qubit is the qubit a controlled gate may change.
Technical
For CX, X acts on the target in the subspace where the control is |1⟩.
Why it matters
In this lab the target is qubit 1. The plus symbol on that wire is the X.
Common misconception
The target is measured first. Measurement order on the diagram is left to right, and both qubits are measured at the end.
Quantum circuit
Plain
A quantum circuit is a sequence of gates, and usually measurements, drawn or written in time order.
Technical
A circuit specifies unitary operations on quantum registers and, when measurements are present, maps results into classical bits.
Why it matters
Composer and Qiskit are two ways to write the same circuit.
Common misconception
A circuit is the quantum computer. The circuit is the program. A simulator or QPU is where you run it.
Quantum register
Plain
A quantum register is the group of qubits a circuit uses.
Technical
In Qiskit, QuantumCircuit(2) creates a quantum register of two qubits, indexed 0 and 1.
Why it matters
Indexing starts at 0. Qubit 0 is the first qubit, not “qubit number 1” in everyday counting.
Common misconception
Register size is the number of shots. Shots are repeats. The register is how many qubits the circuit has.
Classical register
Plain
A classical register is the group of ordinary bits that store measurement results.
Technical
measure_all() adds a classical bit for each qubit and measures into it. The default name in the sampler result is meas.
Why it matters
Counts are tallies of those measurement results.
Common misconception
Classical bits in the circuit are qubits. They are the ordinary record of results.
Measurement
Plain
Measurement asks a qubit for a classical result. You get one basis outcome, not the full list of amplitudes.
Technical
A computational-basis measurement samples a bit string x with probability |αₓ|² and leaves the system in the corresponding basis state.
Why it matters
The lab measures only after you have inspected the statevector of the unmeasured circuit.
Common misconception
Measuring shows the amplitudes. It samples an outcome. Amplitudes are visible in the statevector or in Composer’s state tools, not in one click of a measurement.
Shot
Plain
A shot is one run of the circuit that ends in one measurement result.
Technical
The shots argument tells a sampler how many times to sample the measured circuit.
Why it matters
We use 1024 shots so the histogram is stable enough to talk about, and a challenge asks you to try 100.
Common misconception
More shots make the quantum computer more correct. More shots estimate the same probabilities more smoothly. They do not remove hardware noise.
Counts
Plain
Counts are how many shots produced each bitstring.
Technical
A counts dictionary maps bitstrings to integers. The values sum to the number of shots.
Why it matters
The Python lab prints this dictionary. Keys are bitstrings. Values are tallies.
Common misconception
The biggest count is the statevector. It is the most frequent sample.
Histogram
Plain
A histogram is a bar chart of counts or probabilities for each measured bitstring.
Technical
Composer and Qiskit plot the empirical distribution over computational-basis outcomes.
Why it matters
The Bell lab’s ideal histogram is mostly two bars, 00 and 11.
Common misconception
Every bar you see on hardware is a mistake. Small unexpected bars can be noise. On an ideal simulator they should be absent or negligible.
Simulator
Plain
A simulator is an ordinary computer calculating what an ideal, or sometimes a noisy, quantum circuit would do.
Technical
A statevector simulator stores the 2ⁿ amplitudes and samples them. It is not a quantum processing unit.
Why it matters
We run the ideal simulation first so you know what the circuit is prepared to do.
Common misconception
A simulator result is a quantum-advantage result. It is a classical calculation of the model.
Quantum processing unit or QPU
Q-P-U
Plain
A QPU is the quantum hardware that runs circuits on real qubits.
Technical
IBM Quantum QPUs execute transpiled circuits on superconducting qubits. Results include noise from the device and the environment.
Why it matters
If you later run the same circuit on hardware, you compare the ideal histogram with the QPU histogram. That job uses Open Plan time.
Common misconception
A QPU replaces the need for classical computers. Real projects use both. The Hetionet example is hybrid on purpose.
Noise
Plain
Noise is unwanted disturbance that makes hardware results differ from the ideal state.
Technical
Decoherence, gate errors, readout errors, and crosstalk change probabilities so unexpected bitstrings appear.
Why it matters
A few 01 or 10 counts on hardware, in this Bell lab, are a teaching example of noise, not a new entangled state.
Common misconception
Noise means the device is not quantum. It means the device is a physical system.
Error
Plain
An error is a specific failure: a gate that did not do exactly what the circuit asked, or a measurement that was read wrong.
Technical
Gate error and readout assignment error are quantified in device calibration data. Error mitigation and correction are later topics.
Why it matters
It helps to tell a Python bug, a circuit that does not match the plan, and a hardware error apart from each other.
Common misconception
Any result that is not exactly 512 and 512 is an error. Finite shots fluctuate even in an ideal simulation.
Transpilation
trans-pih-LAY-shun
Plain
Transpilation rewrites your circuit so a particular backend can run it, using the gates and connections that machine actually has.
Technical
The transpiler maps abstract gates onto a backend’s basis gates and coupling map, and may optimize that mapping.
Why it matters
You can ignore it for the ideal two-qubit simulation. It matters as soon as you choose a real backend.
Common misconception
Transpilation changes the problem you meant to solve. It changes the implementation so the hardware can express that circuit.
Backend
Plain
A backend is the simulator or QPU you ask to run a circuit.
Technical
In IBM Quantum and Qiskit Runtime, a backend is a named execution target with its own configuration and, for hardware, calibration.
Why it matters
Composer asks you to choose where to run. The lab starts with an ideal simulation option when it is available.
Common misconception
The backend is your IBM Cloud region. Region and account matter for access. The backend is the compute target.
Qiskit
KIZ-kit
Plain
Qiskit is IBM’s open-source toolkit for writing and running quantum circuits in Python.
Technical
This workshop pins the Qiskit 2.3 line and uses QuantumCircuit, Statevector, and StatevectorSampler.
Why it matters
Composer is the visual tool. Qiskit is the code tool. They describe the same circuit.
Common misconception
Qiskit is the quantum computer. It is software.
IBM Quantum Composer
Plain
Composer is the visual circuit editor in the IBM Quantum Platform. You drag gates onto qubit wires.
Technical
Composer builds OpenQASM-level circuits, shows state information, and can submit jobs to simulators or accessible QPUs.
Why it matters
The first lab is visual so Python does not get in the way of the idea.
Common misconception
Composer is a different kind of quantum computer from Qiskit. It is a different way to write the program.
Quantum kernel
Plain
A quantum kernel is a way of comparing two data points by using a quantum circuit, often so a classical classifier can use that comparison.
Technical
A fidelity quantum kernel estimates overlaps of quantum states prepared by a feature map. QSVC uses such a kernel.
Why it matters
The Hetionet example uses quantum kernels inside QSVC. You will not retrain that model in this workshop.
Common misconception
A quantum kernel replaces the dataset. It is a comparison method inside a larger classical workflow.
Feature map
Plain
A feature map is the circuit pattern that loads classical numbers into a quantum state.
Technical
The Hetionet repository’s best reported quantum path uses a Pauli feature map with two repetitions on 16 features.
Why it matters
Changing the feature map changed the Hetionet scores. The map is a design choice, not a guarantee.
Common misconception
There is one correct feature map for all data. The repository compares maps and reports the result.
Benchmark
Plain
A benchmark is a fair test you can repeat, used to compare methods.
Technical
In the Hetionet project, models are compared on the same prediction task with PR-AUC on held-out or cross-validated splits, as documented in that repository.
Why it matters
Without a benchmark, a quantum experiment is a demo, not evidence.
Common misconception
A benchmark is a leaderboard you must top to justify learning. It is a measurement habit.
Baseline
Plain
A baseline is a strong, simpler method you compare against before claiming a new method helped.
Technical
The Hetionet README reports optimized RandomForest and ExtraTrees classical baselines beside QSVC and stacking ensembles.
Why it matters
The best published number in that README is a hybrid stack, and a classical forest is close behind. That is the honest story.
Common misconception
If a quantum model beats a weak baseline, the problem needs a quantum computer. Beat a serious baseline, and still ask whether the gain is worth the cost.
Hybrid quantum-classical workflow
Plain
A hybrid workflow uses ordinary computers and quantum circuits in one project, each where it fits.
Technical
Hetionet’s documented pipeline embeds a knowledge graph classically, engineers features, trains classical models, evaluates quantum kernels and QSVC, and combines them in a stacking ensemble.
Why it matters
This is the shape of realistic near-term projects, including Fall Fest projects.
Common misconception
Hybrid means the quantum part is decorative. In the documented run, the quantum model is a real component that is measured against baselines.
Quantum advantage
Plain
Quantum advantage would mean a quantum method beats the best practical classical method on a useful task by a margin that matters.
Technical
It is an empirical and problem-specific claim. This workshop does not attempt it, and the Hetionet results are not presented as advantage.
Why it matters
Participants should be able to hear the phrase and not treat this Bell lab as an example of it.
Common misconception
Any circuit that runs on a QPU is quantum advantage. Running is access. Advantage is a comparative claim.
Quantum readiness
Plain
Quantum readiness is knowing why you are exploring quantum computing, who must be involved, what you will learn, and how you will judge progress.
Technical
In this workshop the discussion covers strategy, technology, talent, risk, ecosystems, use-case selection, and workforce pathways, attributed to the leadership themes in Robert Loredo’s Quantum Readiness for Leaders and interpreted for Fall Fest by Quantum Global Group.
Why it matters
It is the point of the Florida Quantum Readiness Challenge onboarding, not a side topic.
Common misconception
Readiness means waiting until you understand all of the physics. It means starting with a clear question and an honest evaluation plan.