Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LinguaQuanta: Towards a Quantum Transpiler Between OpenQASM and Quipper (Extended)

Scott Wesley

TL;DR

LinguaQuanta addresses the interoperability problem between quantum programming languages by building a principled, category-inspired transpiler between Quipper and OpenQASM. The approach uses a modular, UNIX-style pipeline with reflexive-inverse and semantic-preservation properties, enabling stable round-trip translations despite decompositions. The paper develops concrete decomposition techniques, ancilla-management strategies, and shadow-memory instrumentation, and demonstrates the workflow with a Quantum Phase Estimation example to illustrate practical translation and provenance tracking. This work advances quantum software tooling by providing a formal, compositional framework for language interoperability and by outlining concrete future directions for verification and broader language support.

Abstract

As quantum computing evolves, many important questions emerge, such as how best to represent quantum programs, and how to promote interoperability between quantum program analysis tools. These questions arise naturally in the design of quantum transpilers, which translate between quantum programming languages. In this paper, we take a step towards answering these questions by identifying challenges and best practices in quantum transpiler design. We base these recommendations on our experience designing LinguaQuanta, a quantum transpiler between Quipper and OpenQASM. First, we provide categorical specifications for quantum transpilers, which aim to encapsulate the core principles of the UNIX philosophy. We then identify quantum circuit decompositions which we expect to be useful in quantum transpilation. With these foundations in place, we then discuss challenges faced during the implementation of LinguaQuanta, such as ancilla management and stability under round translation. To show that LinguaQuanta works in practice, a short tutorial is given for the example of quantum phase estimation. We conclude with recommendations for the future of LinguaQuanta, and for quantum software development tools more broadly.

LinguaQuanta: Towards a Quantum Transpiler Between OpenQASM and Quipper (Extended)

TL;DR

LinguaQuanta addresses the interoperability problem between quantum programming languages by building a principled, category-inspired transpiler between Quipper and OpenQASM. The approach uses a modular, UNIX-style pipeline with reflexive-inverse and semantic-preservation properties, enabling stable round-trip translations despite decompositions. The paper develops concrete decomposition techniques, ancilla-management strategies, and shadow-memory instrumentation, and demonstrates the workflow with a Quantum Phase Estimation example to illustrate practical translation and provenance tracking. This work advances quantum software tooling by providing a formal, compositional framework for language interoperability and by outlining concrete future directions for verification and broader language support.

Abstract

As quantum computing evolves, many important questions emerge, such as how best to represent quantum programs, and how to promote interoperability between quantum program analysis tools. These questions arise naturally in the design of quantum transpilers, which translate between quantum programming languages. In this paper, we take a step towards answering these questions by identifying challenges and best practices in quantum transpiler design. We base these recommendations on our experience designing LinguaQuanta, a quantum transpiler between Quipper and OpenQASM. First, we provide categorical specifications for quantum transpilers, which aim to encapsulate the core principles of the UNIX philosophy. We then identify quantum circuit decompositions which we expect to be useful in quantum transpilation. With these foundations in place, we then discuss challenges faced during the implementation of LinguaQuanta, such as ancilla management and stability under round translation. To show that LinguaQuanta works in practice, a short tutorial is given for the example of quantum phase estimation. We conclude with recommendations for the future of LinguaQuanta, and for quantum software development tools more broadly.
Paper Structure (41 sections, 17 equations, 25 figures)

This paper contains 41 sections, 17 equations, 25 figures.

Table of Contents

  1. Introduction
  2. A Brief Introduction to Quantum Computation
  3. Unitary Operations on Quantum Systems
  4. Compositionality of Quantum Systems
  5. Circuit Semantics and Compositionality
  6. An Introduction to OpenQASM and Quipper
  7. Pipeline Designs and Categorical Specifications
  8. Decompositions for Quantum Transpilation
  9. Challenges Faced and Proposed Solutions
  10. Approximating the Identity on Round Translation.
  11. Translating OpenQASM Measurements to Quipper.
  12. Typed Wires and Ancillas in OpenQASM.
  13. Ensuring Controls Remain Inlined.
  14. LinguaQuanta by Example
  15. An Example Pipeline in LinguaQuanta
  16. ...and 26 more sections

Figures (25)

  • Figure 1: A collection of quantum gates, and a circuit constructed from those gates.
  • Figure 2: An implementation of QPE in OpenQASM 3 with $3$ digits of accuracy.
  • Figure 3: An implementation of QPE in Quipper with $3$ digits of accuracy.
  • Figure 4: A DFA to infer the state of a Quipper wire. The alphabet consists of Init, Term, and Use. Missing edges indicate one-way transitions to an error state.
  • Figure 5: The translation pipeline implemented in Leblond2023. In this pipeline orange boxes indicate pre-processing, blue boxes indicate post-processing, and green boxes indicate additional steps for legacy support.
  • ...and 20 more figures