BQP $=$ PSPACE

TL;DR

This paper proves that all problems solvable with polynomial space can also be solved efficiently by a quantum computer, establishing that PSPACE equals BQP, with significant implications for quantum algorithms and real-world applications.

Contribution

The paper presents a polynomial-time quantum algorithm for PSPACE-complete problems, showing PSPACE equals BQP, which was previously unknown.

Findings

Quantum algorithms can solve PSPACE-complete problems in polynomial time.

The result implies quantum computers can efficiently perform tasks like quantum channel discrimination.

Potential applications include strategy games and quantum sensing, such as quantum illumination.

Abstract

The complexity class $PSPACE$ includes all computational problems that can be solved by a classical computer with polynomial memory. All $PSPACE$ problems are known to be solvable by a quantum computer too with polynomial memory and are, thus, known to be in $BQPSPACE$. Here, we present a polynomial time quantum algorithm for a $PSPACE$-complete problem, implying that $PSPACE$ is equal to the class $BQP$ of all problems solvable by a quantum computer in polynomial time. In particular, we outline a $BQP$ algorithm for the $PSPACE$-complete problem of evaluating a full binary $NAND$ tree. An existing best of quadratic speedup is achieved using quantum walks for this problem, so that the complexity is still exponential in the problem size. By contrast, we achieve an exponential speedup for the problem, allowing for solving it in polynomial time. There are many real-world applications of our result, such as strategy games like chess or Go. As an example, in quantum sensing, the problem of quantum illumination, that is treated as that of channel discrimination, is $PSPACE$-complete. Our work implies that quantum channel discrimination, and so, quantum illumination, can be performed efficiently by a quantum computer.