Boson Sampling — Photon Interference in a Linear-Optical Network

Boson sampling sends N indistinguishable single photons into distinct input ports of a random mesh of beamsplitters and asks: which pattern of photon counts comes out? This simulator builds a real M-mode linear-optical network from randomly parametrized beamsplitters, computes the exact quantum output distribution via the matrix permanent — the same #P-hard quantity that makes boson sampling believed to be classically intractable — and compares it side-by-side with the distribution classical, distinguishable particles would produce through the identical network. Randomize the mesh, switch between the quantum and classical predictions, and fire single-shot samples to watch multi-photon interference bunch and antibunch the output in ways no classical model reproduces.