Table of Contents
Fetching ...

Understanding the physics of D-Wave annealers: From Schrödinger to Lindblad to Markovian Dynamics

Vrinda Mehta, Hans De Raedt, Kristel Michielsen, Fengping Jin

TL;DR

The paper investigates whether D-Wave annealers operate under genuine quantum dynamics or can be explained by classical or Markovian processes. By studying 1- and 2-spin problems under standard and fast annealing, and by comparing experimental data to Schrödinger, Lindblad, and Markovian models, it shows that long annealing drives sampling toward Gibbs-like equilibrium, with GKSL-Lindblad and Markovian frameworks able to reproduce much of the observed behavior. A notable dips-and-bumps feature in standard annealing and the limited coherence observed during fast annealing suggest a nuanced interplay between quantum and classical dynamics. These results clarify the role of quantum effects in D-Wave devices and inform strategies for leveraging or mitigating environment-induced dynamics in optimization tasks.

Abstract

Understanding the physical nature of the D-Wave annealers remains a subject of active investigation. In this study, we analyze the sampling behavior of these systems and explore whether their results can be replicated using quantum and Markovian models. Employing the standard and the fast annealing protocols, we observe that the D-Wave annealers sample states with frequencies matching the Gibbs distribution for sufficiently long annealing times. Using Bloch equation simulations for single-qubit problems and Lindblad and Markovian master equations for two-qubit systems, we compare experimental data with theoretical predictions. Our results provide insights into the role of quantum mechanics in these devices.

Understanding the physics of D-Wave annealers: From Schrödinger to Lindblad to Markovian Dynamics

TL;DR

The paper investigates whether D-Wave annealers operate under genuine quantum dynamics or can be explained by classical or Markovian processes. By studying 1- and 2-spin problems under standard and fast annealing, and by comparing experimental data to Schrödinger, Lindblad, and Markovian models, it shows that long annealing drives sampling toward Gibbs-like equilibrium, with GKSL-Lindblad and Markovian frameworks able to reproduce much of the observed behavior. A notable dips-and-bumps feature in standard annealing and the limited coherence observed during fast annealing suggest a nuanced interplay between quantum and classical dynamics. These results clarify the role of quantum effects in D-Wave devices and inform strategies for leveraging or mitigating environment-induced dynamics in optimization tasks.

Abstract

Understanding the physical nature of the D-Wave annealers remains a subject of active investigation. In this study, we analyze the sampling behavior of these systems and explore whether their results can be replicated using quantum and Markovian models. Employing the standard and the fast annealing protocols, we observe that the D-Wave annealers sample states with frequencies matching the Gibbs distribution for sufficiently long annealing times. Using Bloch equation simulations for single-qubit problems and Lindblad and Markovian master equations for two-qubit systems, we compare experimental data with theoretical predictions. Our results provide insights into the role of quantum mechanics in these devices.

Paper Structure

This paper contains 19 sections, 33 equations, 16 figures, 1 table.

Figures (16)

  • Figure 1: (Color online) D-Wave data obtained using the standard annealing protocol for a 1-spin problem with (a) $h_1=0.1$ and (b) $h_1=0.2$.
  • Figure 2: (Color online) D-Wave data obtained using the standard annealing protocol for a 2-spin problem instances with (a) 2S1, (b) 2S2, and (c) 2S3.
  • Figure 3: (Color online) D-Wave data for the ferromagnetic spin chain of length $N=10,\ldots,1000$. (a) Absolute value of the average energy $\langle E(t_a)\rangle$ as a function of annealing time $t_a$ and (b) the average energy at $t_a=1950~\mu$s as a function of chain length $N$.
  • Figure 4: (Color online) D-Wave data obtained using the fast annealing protocol for 1-spin problem with $h_1=0.25$.
  • Figure 5: (Color online) D-Wave data obtained using the fast annealing protocol for a 2-spin problem instances with (a) 2S1, (b) 2S2, and (c) 2S3.
  • ...and 11 more figures