RQC Seminar

21st RQC Seminar

• Speaker

  Mr. Zane Marius Rossi
  (PhD Student, MIT Quanta Group — Cambridge, MA)

• Date

  14:00-15:00 (JST), November 22, 2022 (Tuesday)

• Venue

  Hybrid (Zoom / Wako Welfare and Conf. 2F Large Meeting Room RIKEN Wako branch)

• Title

  Unifying and improving quantum algorithms

• Inquiries

  rqc_info[at]ml.riken.jp

Abstract
Recent work has shown that quantum signal processing (QSP) and its multiple-qubit lifted version, quantum singular value transformation (QSVT), unify and improve the presentation of most quantum algorithms (from unstructured search to factoring to Hamiltonian simulation). QSP/QSVT characterize the ability, by special alternating circuit ansätze, to obliviously transform the singular values of sub-blocks of unitary matrices by polynomial functions; these algorithms are numerically stable, flexible, and analytically well-understood. Moreover, their applicability to a wide class of linear algebraic problems has made the conversion of known quantum algorithms to the QSVT framework a common mode of research. A recent question is whether QSP/QSVT serve primarily to explain existing quantum algorithms, or whether they also have the power to suggest new paths toward quantum advantage. Here we present recent substantial elaborations on these algorithms, including the cases of multiple unitary oracles and the recursive embedding of QSP protocols—surprisingly, many formal properties of QSP/QSVT can be analogized to and proven in these expanded settings.