Speaker: Stephen Jordan, Google Quantum AI

 

 

Abstract: In this talk I will describe Decoded Quantum Interferometry (DQI), a quantum algorithm for reducing classical optimization problems to classical decoding problems by exploiting structure in the Fourier spectrum of the objective function. (See: https://arxiv.org/abs/2408.08292.) For a regression problem called optimal polynomial intersection, which has been previously studied in the contexts of coding theory and cryptanalysis, DQI achieves an exponential quantum speedup over all classical algorithms we are aware of. We also investigate the application of DQI to average-case instances of max-k-XORSAT. DQI reduces max-k-XORSAT to decoding LDPC codes, which can be achieved using powerful classical algorithms such as belief propagation. In this setting we identify a family of max-XORSAT instances where DQI achieves a better approximation ratio than simulated annealing, although not better than specialized classical algorithms tailored to those instances. The recent quantum query complexity speedup of Yamakawa and Zhandry can also be obtained as a special case of DQI. This is joint work with Noah Shutty, Mary Wootters, Adam Zalcman, Alexander Schmidhuber, Robbie King, Sergei V. Isakov, and Ryan Babbush.

 

Bio: Stephen Jordan is a researcher in Google Quantum AI. The main focus of his research is quantum algorithms. He obtained his PhD in physics in 2008 from MIT. Prior to joining Google he worked at NIST and Microsoft and was a QuICS fellow at the University of Maryland. Additionally, he is the author and maintainer of the quantum algorithm zoo, an online repository of quantum algorithms.