Three Paradoxes of Quantum Information
Subhash C. Kak
Abstract
Three paradoxes of quantum information are presented. These paradoxes concern the final state obtained by applying unitary transformations on a single qubit that belongs to an entangled pair. We show that certain operations appear to yield the same final state for different starting states. Given entangled particles, it is reasonable to apply unitary transformations on any single particle, obtaining the usual predictions for unentangled qubits. The requirements of invertibility mean that transformations on different initial states will lead to different final states. One may view the situation of the entangled particles from two perspectives. In the first, the state of the two particles is considered jointly. In the second perspective, one may consider each particle separately, without the knowledge that it is entangled with another particle. From this perspective, consistency demands that unitary transformations on any one particle will provide the same results as when the particles are unentangled. The paradoxes presented below follow from elementary operations on entangled particles. Some readers may not consider them as true paradoxes, taking them to arise from the manner of the derivation as an artifact of the representation. But the representation used here is the one commonly accepted in quantum information science[1]. The three paradoxes are related and may be considered versions of the same paradox; here I give them separately to highlight different aspects of the underlying problem. Paradox 1 Consider the entangled state 1 √ 2 (|00 〉 + |11〉). Both particles (qubits) have equal probability of being 0 or 1. We may assume that the particles have traveled apart so that one may operate on each separately, without influencing the other. Consider now the unitary transformation H = 1 []