Skip to content
Quantum.world

The Signal

Novikau and Joseph Build a Quantum Circuit for Dissipation

Ivan Novikau and Ilon Joseph describe a quantum circuit for modeling dissipative evolution as a weighted sum of Hamiltonian simulations. They test the method on an advection-diffusion equation using a digital emulator of a fault-tolerant quantum computer.

Novikau and Joseph Build a Quantum Circuit for Dissipation

Quantum circuits use unitary operations, but many equations used to model physical processes describe dissipative, non-unitary change. Ivan Novikau and Ilon Joseph’s new algorithm tackles that mismatch with a circuit that represents dissipative evolution as a weighted sum of Hamiltonian simulations. Their paper appeared in the Journal of Computational Physics on November 1, 2026, and tests the method on an advection-diffusion equation.

Source sciencedirect.com

Turning dissipative evolution into Hamiltonian simulations

The authors build on the Linear Combination of Hamiltonian Simulations (LCHS) method. Rather than implementing a dissipative operator directly, LCHS approximates it by combining unitary Hamiltonian evolutions at different time scales. The approach lets a quantum circuit model non-unitary dynamics through operations that fit the circuit framework.

Novikau and Joseph focus on making that combination explicit and more efficient to encode. Their coordinate transformation turns the dependence on the LCHS summation index into a trigonometric function, which simplifies the circuit’s block-encoding. The paper connects the same construction, when computed classically, to a Fejér-Clenshaw-Curtis quadrature formula.

One QSP circuit handles many terms

The resulting design uses a single Quantum Signal Processing (QSP) circuit to perform an exponential number of Hamiltonian simulations. Its selector—the circuit component that selects among terms in the sum—scales logarithmically with the number of terms and linearly with simulated time. The authors say their convergence analysis makes the circuit more efficient than other recent LCHS circuits.

“The resulting LCHS circuit has high success probability and the selector scales logarithmically with the number of terms in the LCHS sum and linearly with time.”

Ivan Novikau and Ilon Joseph, the paper’s authors

That scaling describes the selector, not a complete hardware resource estimate for every problem. The paper’s abstract does not give a universal gate count or a measured speedup over classical solvers. Practical cost will also depend on how a particular differential equation is encoded and what output a user needs to recover.

Advection-diffusion supplies the test case

To check the circuit and its scaling, the researchers simulate an advection-diffusion equation on a digital emulator of a fault-tolerant quantum computer. The equation describes how a quantity moves through a medium while spreading, making it a direct test of the method’s intended use for dissipative initial-value problems. The paper reports this as a simulation, not a run on a physical quantum processor.

The authors say the technique could also apply to other non-unitary initial-value problems. Their examples include the Liouville equation and linear embeddings of nonlinear systems, such as Koopman-von Neumann and Carleman formulations. Those extensions are potential applications of the algorithm; the reported test in this paper is the advection-diffusion equation.

The emulator result is not a hardware demonstration

The distinction matters because the method’s central result is an implementation strategy, not an experimental demonstration of quantum advantage. An emulator can verify circuit behavior and probe resource scaling, but it does not establish that a quantum processor can solve a useful dissipative problem faster or at lower cost than a classical solver. The paper presents the circuit as a route toward simulating a broad class of equations, while leaving hardware performance and application-level advantage untested.

For now, the concrete result is a specified LCHS circuit with a logarithmically scaling selector and an emulator-based advection-diffusion test. Novikau and Joseph’s contribution is to make the summation over Hamiltonian evolutions more compact to encode, while keeping the time dependence linear in the selector.

Source

  1. ScienceDirect (Elsevier)

Related stories

Explore

More articles