Differentiable quantum architecture search

Abstract Quantum architecture search (QAS) is the process of automating engineering quantum circuits. It has been desired to construct a powerful and general QAS platform which can significantly accelerate current efforts identify advantages error-prone depth-limited circuits in NISQ era. Hereby, we propose framework differentiable (DQAS), enables automated designs an end-to-end fashion. We present several examples circuit design problems demonstrate power DQAS. For instance, unitary operations are decomposed into gates, noisy re-designed improve accuracy, layouts for approximation optimization algorithm automatically discovered upgraded combinatorial problems. These results not only manifest vast potential DQAS being essential tool application developments, but also interesting research topic from theoretical perspective as it draws inspirations newly emerging interdisciplinary paradigms programming, probabilistic programming.