Optimizing a quantum circuit is hard because it requires exploring a vast space of functionally equivalent circuits, produced by applying local circuit rewrites such as gate cancellation and commutation. Each additional rewrite can exponentially expand the space of equivalent circuits, so existing optimizers can only explore a tiny fraction of this space and often produce suboptimal results. We present Quasar, a new quantum circuit optimizer that can generate a step-limited optimal circuit: given a bound on rewrite iterations, it returns the lowest-cost circuit reachable within that bound. To achieve this, Quasar constructs two complementary e-graphs for the quantum circuit’s graph and sequence representations. Quasar then infers an atomic rewrite set for application on both e-graphs, which improves optimization performance by orders of magnitude. Finally, Quasar employs a series of new optimization techniques to ensure both soundness and scalability of equality saturation on quantum circuits. Across standard benchmarks, Quasar achieves geometric-mean reductions of 20.2% in 2-qubit gate count, 33.5% in total gate count, and 24.2% in circuit depth, and a 21.4% fidelity improvement over unoptimized circuits, outperforming or matching state-of-the-art rewrite-based optimizers on 75%, 92%, 62%, and 80% of circuits, respectively.
Yang et al. (Mon,) studied this question.