We propose and experimentally demonstrate an approximate universal-NOT (UNOT) operation that is robust against operational errors. In our proposal, the UNOT operation is composed of stochastic unitary operations represented by the vertices of regular ...
We propose and experimentally demonstrate an approximate universal-NOT (UNOT) operation that is robust against operational errors. In our proposal, the UNOT operation is composed of stochastic unitary operations represented by the vertices of regular polyhedrons. The operation is designed to be robust against random operational errors by increasing the number of unitary operations (i.e., reference axes). Remarkably, no increase in the total number of measurements nor additional resources are required to perform the UNOT operation. Our method can be applied in general to reduce operational errors to an arbitrary degree of precision when approximating any antiunitary operation in a stochastic manner.