Quantum evaluation of fixed-point Gaussian functions
Follow the latest progress by email.
Can a reversible circuit evaluate a fixed-point Gaussian function with fewer Toffoli gates and peak qubits while meeting its prescribed approximation error?
Why it matters and what counts as progress
Why it matters: Gaussian evaluation is a numerical primitive whose reversible arithmetic can dominate a larger quantum routine.
What counts as progress: A cheaper circuit over the same promised domain and precision, retaining a worst-case error guarantee such as the reported 9.99×10⁻⁸ benchmark.
How the problem developed
- 15 Oct 2026 · Circuit optimization [1]
Andi Gu, Alek H. Kemeny, Susanne F. Yelin and Mikhail D. Lukin, using LLM-guided search Reported a fixed-point Gaussian circuit with 1,583 Toffolis and 139 qubits, versus a tuned Häner–Roetteler–Svore baseline of 24,000 Toffolis and 362 qubits (39× lower Toffoli–qubit product).
AI contribution
Discovering state-of-the-art quantum circuits with large language models
Gu, Kemeny, Yelin and Lukin report lower Toffoli and peak-qubit counts in seven specified circuit benchmarks, plus a new prime-field modular-inversion architecture. For fixed-point Gaussian evaluation, their stated benchmark falls from about 24,000 Toffolis and 362 qubits to 1,583 Toffolis and 139 qubits; the reported worst-case error is 9.99×10⁻⁸.
Sources
Website timeline
Publication dates and editorial changes.
LLM-discovered quantum circuits reduce Toffoli and qubit costs