Why is simulating quantum systems classically intractable, but a quantum computer can do it efficiently?
Think about your answer, then reveal below.
Model answer: Classical simulation requires exponentially many parameters to describe a quantum state: a system of n qubits requires 2^n complex amplitudes to fully specify. Classically simulating n=100 qubits requires storing 2^100 ≈ 10^30 complex numbers, which is impossible. A quantum computer directly encodes the n-qubit state, using only n qubits. Time evolution of the quantum state is also exponentially hard classically (each timestep multiplies by the 2^n × 2^n Hamiltonian matrix), but on a quantum computer, applying gates directly evolves the state in time linear in the circuit depth. This exponential advantage is the reason quantum simulation is one of the most promising near-term applications.
The exponential cost of classical simulation is the key motivation for quantum computing in chemistry and materials science. This advantage is relative (not absolute), but sufficient to make otherwise intractable simulations possible.
Question 2 Multiple Choice
The Trotter-Suzuki formula breaks e^{i H t} into products of e^{i H_i t/k} for k segments. What is the error when using a finite k?
AError is O(t^2 / k)
BError is O(t^3 / k^2)
CError is O(1 / k^2), independent of t
DTrotter-Suzuki is exact; there is no error
The first-order Trotter formula has error O(t^3 / k^2). This is derived from the Baker-Campbell-Hausdorff formula; successive Trotter applications compound errors. Practical implementations use higher-order Suzuki formulas (4th order, 6th order) with better error scaling. The dependence on t^3 reflects that longer simulations accumulate more error. For practical quantum simulations, choosing k (number of segments) is a crucial trade-off: larger k reduces error but requires more gates.
Question 3 True / False
Variational approaches (VQE, QAOA) for quantum simulation use classical-quantum feedback loops. What is the advantage over direct Hamiltonian simulation?
TTrue
FFalse
Answer: True
Variational approaches run on near-term noisy devices by keeping circuit depth shallow, avoiding the accumulation of gate errors from many gates. Direct Hamiltonian simulation requires accurately implementing U = e^{i H t}, which for long times t requires many gates and deep circuits. Variational methods use a shallow ansatz (parameterized circuit), estimate its energy via measurements, and classically optimize parameters. This trades off the rigor of simulating the true Hamiltonian for practicality on NISQ devices. The trade-off is worthwhile when you only need the ground state energy or low-lying excitations, not the full time-evolved state.