Quantum Simulations Bypass Complex Calculations with Novel Technique

A new variational method for modelling the behaviour of interacting, open quantum spin systems is presented by Jacopo Tosca of the Paris Cité University and colleagues. The technique is based on the spin phase-space representation, variationally targeting the Husimi-$Q$ function with a novel ansatz utilising negative mixture coefficients to accurately represent quantum correlations. Derived from the Dirac-Frenkel variational principle, the approach avoids computationally expensive Monte Carlo sampling through analytical exploitation of the ansatz structure. Accurate simulations of the dynamics and non-equilibrium steady states of the transverse-field quantum Ising model are achieved, and it shows efficient scaling to large two-dimensional lattices, exceeding the capabilities of existing methods.

Variational dynamics of open quantum spin systems on two-dimensional lattices

The new variational method accurately simulates the dynamics of interacting, open quantum spin systems on two-dimensional lattices, achieving accuracy comparable to exact diagonalization for systems up to 8×8, a scale previously inaccessible to most techniques. Exact diagonalization becomes computationally prohibitive beyond approximately 16 sites, severely limiting the study of larger, more complex quantum phenomena. By avoiding Monte Carlo sampling and utilising a phase-space representation with negative mixture coefficients, the method efficiently calculates equations of motion, capturing both quantum dynamics and non-equilibrium steady states with remarkable precision.

The transverse-field quantum Ising model served as a test case, confirming the method’s efficacy and scalability, and opening new avenues for modelling complex quantum materials and devices. Simulating the dynamics of a one-dimensional transverse-field Ising model with 16 spins required approximately one minute on a standard desktop computer to model evolution up to 31 units of time, and achieved precise agreement with exact Monte Carlo wave function simulations. A two-dimensional 3×3 lattice demonstrated accurate real-time dynamics and steady states using only 280 variational parameters, computed in roughly ten seconds; this contrasts sharply with the difficulties faced by neural network approaches in similar two-dimensional systems. Scaling to a 8×8 lattice, the simulation accurately captured the time evolution of three spin components, x, y, and z, and showed systematically improved convergence as the number of coherent states increased from two to six, resulting in parameter counts ranging from 386 to 1158. The system was mapped onto this ‘phase space’ instead of directly calculating the complex evolution of each quantum particle, considerably simplifying the calculations.

Husimi-Q function variational ansatz for open spin systems

Representing quantum states using the Husimi-$Q$ function underpins this advance, offering a way of representing quantum states as a probability distribution, similar to how a blurred photograph captures the general shape of a moving object. This representation then allowed the construction of a variational ansatz, essentially a carefully chosen approximation, using multidimensional mixtures of spin-coherent states; these states behave in a relatively predictable way, akin to a laser beam which maintains a consistent direction and intensity.

Multidimensional mixtures of spin-coherent states are employed, allowing negative coefficients to accurately capture quantum correlations and avoiding the computational demands of Monte Carlo sampling. It successfully simulates the dynamics of the transverse-field quantum Ising model and scales efficiently to larger two-dimensional lattices, overcoming limitations of existing techniques. The ability to model quantum systems without reliance on computationally intensive methods offers a significant advantage in exploring complex phenomena.

Simulating quantum spin systems without Monte Carlo methods using a new computational approach

A new computational technique has been devised to model the behaviour of complex quantum systems, offering a potential route to designing future technologies. It accurately simulates quantum dynamics and steady states, avoiding the intensive calculations of Monte Carlo sampling, although it is currently limited to the transverse-field quantum Ising model. Generalising the approach to encompass a broader range of interacting quantum spin systems, each with unique characteristics and complexities, remains an open question.

Despite being presently confined to the transverse-field quantum Ising model, this new technique represents strong progress in the field. It offers a way to bypass computationally expensive Monte Carlo methods, crucial for simulating complex quantum systems underpinning potential advances in materials science and computing. Accurately modelling dynamics and steady states, even within these limitations, establishes a foundation for future generalisations and broader applications across diverse quantum spin systems.

This new technique establishes a viable phase-space approach for accurately modelling interacting quantum spin systems. Achieving efficient simulations, particularly in two dimensions, marks a considerable advance in understanding quantum phenomena and their potential technological applications. By avoiding computationally intensive Monte Carlo methods, it is now possible to explore larger, more intricate quantum systems than previously possible. Further investigation is required to extend the technique beyond the transverse-field quantum Ising model, raising questions regarding its adaptability to diverse quantum systems exhibiting varied interactions and complexities.

The researchers developed a new computational method to simulate interacting open quantum spin systems without using Monte Carlo sampling. This is important because it allows for more efficient modelling of complex quantum dynamics and steady states, particularly for larger two-dimensional lattices. The technique accurately captured the behaviour of the transverse-field quantum Ising model, demonstrating its potential for understanding quantum phenomena. The authors note that extending this approach to a wider range of quantum spin systems is a key area for future work.