Observation of disorder-free localization using a (2+1)D lattice gauge theory on a quantum processor

arXiv:2410.06557 · quant-ph, cond-mat.dis-nn, cond-mat.str-el, hep-lat · Submitted 2024-10-09 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: I'm Kai, and with me are Mira and Lev, guest researcher.

Mira: Today's paper: "Observation of disorder-free localization using a (2+1)D lattice gauge theory on a quantum processor".

Kai: This research investigates disorder-free localization (DFL) in quantum many-body systems by leveraging translationally invariant evolutions of lattice gauge theory (LGT) Hamiltonians on a quantum processor.

Mira: First, who's behind it and why it matters.

Title and authors: Kai: So, let's talk about the title and the authors of this paper, "Observation of disorder-free localization using a (two plusone)D lattice gauge theory on a quantum processor." The paper is essentially proposing to use a specific type of quantum hardware to observe how localization behaves when there's no explicit disorder present in certain lattice gauge theories.

Mira: I think the title immediately signals that they are focusing on two key areas: localization and the use of lattice gauge theory. This suggests their approach isn't just about standard disordered systems but using this specific theoretical tool to probe a more fundamental aspect of quantum mechanics.

Lev: From my perspective, having them use a lattice gauge theory framework means we’re dealing with something that has inherent symmetries that might simplify the Hamiltonian, which is good for analysis but hard for implementation on current devices.

Kai: That's right; the LGT structure provides those underlying symmetries that allow them to define what they call "superpositions over all gauge sectors," which is a key concept we need to understand better.

Mira: And that concept allows them to interpret translationally invariant states as superpositions over all gauge sectors, which is a powerful way to handle the complexity of disorder in these simulations.

Lev: If they can successfully prepare those states on the quantum processor, it means they’ve solved a significant initial hurdle in translating abstract theoretical concepts into physical qubit operations.

Kai: It’s about showing that we can construct these complex initial states using circuits, as shown by their schematic of preparing translationally invariant states of thirty-eight qubits with an energy perturbation at the center link (j = ten).

Mira: And that specific setup, where they initialize the center gauge qubit in a direction opposite to the rest on the XZ-plane of the Bloch sphere, seems like a very carefully chosen starting point.

Lev: The complexity of setting up those precise initial conditions is exactly what we worry about when you try to run this on actual hardware and minimize gate depth for error mitigation.

Kai: It’s a lot of delicate control over the qubits, but they demonstrate that this level of control is achievable, which opens doors for more complex physics simulations.

Mira: Overall, the authors are connecting deep theoretical ideas about gauge sectors to tangible quantum circuits, showing how these concepts translate into observable dynamics on a quantum processor.

Lev: And if they can do that reliably, it sets a precedent for other researchers to use this LGT approach as a starting point for studying localization in more realistic physical systems.

Kai: So, the title really captures the intersection of theoretical physics and cutting-edge quantum hardware implementation. The paper is about showing that we can build simulations of disordered systems in a controlled way using these novel quantum techniques.

Mira: It’s about bridging that gap between abstract theory and what's actually being built on the hardware, which is where a lot of the real excitement lies for condensed matter theorists.

The paper's summary: Kai: Moving on to the actual summary of "Observation of disorder-free localization using a (two plusone)D lattice gauge theory on a quantum processor," they essentially explain that in their translationally invariant LGT, they observe localization in the absence of explicit disorder when the system is evolved without any perturbations.

Mira: That's right; the core finding is that perturbations fail to diffuse even during fully disorder-free evolution and initial states in one and two dimensions, which goes against conventional expectations for how energy should spread.

Lev: That persistence of localization on large timescales is what makes this result interesting because it suggests a non-trivial dynamical feature rather than just a simple artifact of finite simulation time.

Kai: And they then immediately pivot to the fact that while R´enyi entropy measurements show that superposition-prepared states are fundamentally different from those obtained by direct disorder sampling, which points to the importance of their initial state preparation method.

Mira: That distinction is critical because it shows that the way they prepare these states matters for what we measure, and they also propose a new algorithm with a polynomial speedup in sampling disorder configurations as a major methodological contribution.

Lev: The paper claims this polynomial speedup is achievable through the Grover search algorithm applied to estimate single-qubit Pauli observables, which is a significant claim for practical implementation.

Kai: So, it’s not just an observation of localization, but also a concrete proposal for how to tackle the hard problem of sampling disorder configurations efficiently in many-body localization studies.

Mira: And they also mention that the Floquet unitary commutes with certain conserved operators, Gˆj = ˆσXj Yk∈N(j) Xˆj,k, which is a disorder-free linear exponential in N.

Lev: That commutation relation suggests a strong underlying symmetry that keeps things stable over time even when dealing with the non-integrable nature of the LGT model.

Kai: And they show that they can verify these results using MPS simulations for 2D systems and Exact Diagonalization for 1D systems, confirming the experimental results on a computational level.

Mira: The overall summary is that they’ve provided both an observation of localization and a concrete method to sample disorder efficiently, linking these two ideas together through the framework of translationally invariant states in LGT.

Lev: It sounds like they’ve done a lot of work connecting the dots between the theoretical concepts and what we see on the quantum processor.

The paper's improvements: Kai: Now, let's look at the specific improvements suggested by this paper, and they focus heavily on leveraging this superposition technique to achieve that polynomial speedup in sampling disorder configurations, which is a major advancement over traditional methods.

Mira: I think the most impactful improvement is definitely the proposal of using Grover's algorithm for estimating Pauli observables with high accuracy in a polynomial number of applications to estimate disorder-averaged expectation values, which is a significant methodological step.

Lev: From an error correction standpoint, that means we need to focus our efforts on optimizing those unitary operations because if we can achieve this speedup, the required resources for simulating complex MBL systems drops dramatically.

Kai: And they also suggest developing a robust framework for probing and distinguishing between different types of non-ergodic behavior like disorder-free localization versus many-body localization in lattice gauge theories, which is important for classification.

Mira: That framework helps us classify the underlying physical mechanism causing transport suppression or localization in complex material science problems, which is a big step for applying these findings beyond pure theory.

Lev: If they can successfully map these LGT results onto physical models relevant to condensed matter physics and topological phases, that gives us more concrete targets for error correction research.

Kai: And finally, there’s the methodology for performing high-fidelity quantum simulations of complex, interacting Hamiltonians by employing Matrix Product States with optimized bond dimensions to simulate larger system sizes and longer time scales than currently feasible.

Mira: That capability is essential because it allows us to test new quantum algorithms and simulate more realistic physical models that involve strong interactions and disorder, which is crucial for testing the limits of what we can model.

Lev: And I think the ability to run these simulations with optimized bond dimensions gives us a realistic look at how much resources are actually needed before we can predict system breakdown.

Kai: So, in summary, the improvements center on making the simulation more efficient through smarter sampling and improving our ability to accurately characterize these non-ergodic phases using this new framework.

Mira: It’s a synthesis of a new sampling technique and improved characterization tools that makes this work much more applicable to studying real physical phenomena.

Conclusion: Kai: So, to wrap up on the paper "Observation of disorder-free localization using a (two plusone)D lattice gauge theory on a quantum processor," the main implication is that we have established that energy excitations can remain localized even in the presence of spatial disorder under these specific conditions.

Mira: That finding, combined with their new sampling algorithm and the polynomial speedup for estimating disorder-averaged expectation values, offers a new way to study disordered systems that doesn't require explicit averaging over disorder realizations.

Lev: For me, it means we have a better conceptual tool to handle the complexity of MBL studies without getting bogged down in brute-force sampling issues.

Kai: I think the real excitement is in seeing how this could lead to new material designs and testing of these fundamental quantum principles on actual hardware.

Mira: It’s a significant step forward because it provides a clearer theoretical map for understanding why certain excitations remain localized even when the underlying system has spatial disorder, informing our models for complex quantum materials.

Lev: I think the polynomial speedup in sampling is the most practical win we have right now, making this work more feasible for real-world applications.

Kai: I think we've got a lot of exciting ground to cover as we look at how these LGT results can translate into new experimental setups and algorithms.

Mira: It’s certainly a significant contribution to the field because it provides a clearer path forward for studying disorder in many-body systems without needing that explicit averaging.

Lev: I just think the whole picture is very promising for what we can achieve with this kind of research.

Google Quantum AI and Collaborators

Google Quantum AI

quant-ph, cond-mat.dis-nn, cond-mat.str-el, hep-lat

Submitted: 2024-10-09

Updated: 2025-07-06

Journal ref: Science 393, 71-75 (2026)

DOI: 10.1126/science.adr9680

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 82/100

The gist: This research investigates disorder-free localization (DFL) in quantum many-body systems by leveraging translationally invariant evolutions of lattice gauge theory (LGT) Hamiltonians on a quantum

Key concepts

Disorder-Free Localization (DFL): This refers to the observation that in a translationally invariant lattice gauge theory, perturbations fail to diffuse energy even when there is no explicit disorder present. This persistence of localization on large timescales suggests a non-trivial dynamical feature in these systems.
Lattice Gauge Theory (LGT): LGT provides inherent symmetries to the Hamiltonian, which simplifies analysis. The framework allows researchers to define 'superpositions over all gauge sectors,' offering a powerful way to handle the complexity associated with disorder simulations.
Polynomial Speedup in Sampling: The paper proposes a new algorithm using Grover's search for estimating single-qubit Pauli observables. This method offers a polynomial speedup for sampling disorder configurations, which is considered a major methodological contribution over traditional methods.

Terminology

Fidelity — 29× in this paper · explained in 2 episodes
Entanglement — 26× in this paper · explained in 3 episodes
Hilbert Space — 9× in this paper · explained in 2 episodes
Matrix Product States — 7× in this paper
Exact Diagonalization — 6× in this paper
Quantum Circuits — 6× in this paper · explained in 2 episodes
Accuracy — 5× in this paper · explained in 1 episode
Entanglement Entropy — 5× in this paper · explained in 2 episodes
Matrix Product States (MPS) — 5× in this paper · explained in 2 episodes
Ancillary Qubits — 4× in this paper · explained in 1 episode
Core — 4× in this paper · explained in 2 episodes
Ancilla Qubits — 3× in this paper · explained in 1 episode
Benchmarking — 3× in this paper · explained in 3 episodes
Trace — 3× in this paper · explained in 2 episodes
Alignment — 1× in this paper · explained in 1 episode
Area Law — 1× in this paper · explained in 1 episode
Conservation Laws — 1× in this paper · explained in 3 episodes
European Research Council — 1× in this paper
Frustration — 1× in this paper · explained in 2 episodes
Hamming Distance — 1× in this paper · explained in 2 episodes
LGT Lattice Gauge Theory — 1× in this paper
Magnetic Frustration — 1× in this paper · explained in 2 episodes
Phase Transitions — 1× in this paper · explained in 2 episodes
Topological Order — 1× in this paper · explained in 2 episodes
Unitary Dynamics — 1× in this paper · explained in 2 episodes
Verification — 1× in this paper · explained in 2 episodes

Summary

This research investigates disorder-free localization (DFL) in quantum many-body systems by leveraging translationally invariant evolutions of lattice gauge theory (LGT) Hamiltonians on a quantum processor. This work is significant because it demonstrates that energy excitations can remain localized even in the presence of spatial disorder, challenging conventional wisdom regarding localization and providing a new framework for studying disordered systems without requiring explicit averaging over disorder realizations.

The Theoretical Framework: Lattice Gauge Theory and Disorder

The study utilizes a translationally invariant lattice gauge theory (LGT) Hamiltonian in (1+1)D and (2+1)D, specifically focusing on the dynamics generated by the Trotterized evolution of this Hamiltonian. The core idea is that translationally invariant states can be interpreted as superpositions over all gauge sectors. In the absence of disorder, localization is observed in one and two dimensions; perturbations fail to diffuse despite fully disorder-free evolution. This observation suggests that localization can persist on very large timescales even with interactions, potentially leading to a non-equilibrium many-body localized (MBL) phase when increasing disorder strength.

Quantum Simulation Protocol

The researchers implement a protocol that uses quantum superposition to effectively sample disorder configurations, addressing the challenge of sampling disorder realizations. This is achieved by augmenting the system with ancillary qubits and mapping the binary disorder coupling term, denoted as Hˆdis = Pj gjDˆj, to an ancilla operator: Hˆ Q dis = Pj σXj Dˆj. Crucially, they show that preparing all ancillas in the state 00 · · · 0⟩ results in a superposition over all disorder configurations with equal weights. The system is then evolved under Hˆ Q = Hˆ ord ⊗ 1ancilla + Hˆ Q dis, and measuring expectation values after time-evolution is equivalent to the disorder-average defined in Eq. (1).

Symmetry and Sector Identification

The localization mechanism is deeply tied to the extensive local symmetries of the LGT Hamiltonian. The researchers show that a local unitary transformation exists where symmetry generators appear as couplings, allowing them to identify disorder-free sectors. These translationally invariant states correspond to superpositions over all symmetry sectors (i.e., background potentials). In most sectors, the dynamics take place in a disordered background, which provides an explanation for the localization of energy. Furthermore, they note that the Floquet unitary commutes with each conserved operator Gˆj = ˆσXj Yk∈N(j) Xˆj,k (Eq. 5), which is a disorder-free linear exponential in N.

Observables and Localization Signatures

The paper examines several observables to characterize the dynamics, including:

Energy excitations: They observe that for certain nearly translationally invariant initial conditions in both one and two dimensions, energy excitations remain localized.
Gauge polarization, matter polarization, and interaction per link: These are measured as expectation values of terms within the Floquet unitary (Fig. 1C-E).
Second R´enyi entropy: They measure this non-linear quantity to distinguish between initial states formed by superpositions over all disorder configurations and those obtained by direct disorder sampling, noting that the entropy of the superposition state is larger compared to the disorder-averaged entropy of a MBL system.

Sampling Advantage via Grover's Algorithm

The work proposes an algorithm leveraging quantum superposition for efficient sampling. By constructing a unitary UˆLGT such that ⟨0Uˆ † LGTOˆUˆLGT0⟩ = X D Pr(D) O E D (Eq. 30), they show that the Grover search algorithm can estimate single-qubit Pauli observables to accuracy ∼ ϵ in ∼ 1/ϵ applications of UˆLGT and O(1) experiments, offering a polynomial speedup over naive sampling, which would otherwise require O(1/ϵ 2) applications of UˆD. This advantage is key for studying rare events crucial to understanding the MBL phase.

Numerical Verification via MPS Simulations

To verify the experimental results, they employ Matrix Product States (MPS) simulations for 2D systems and Exact Diagonalization (ED) for 1D systems. The cost scaling analysis shows that while entanglement grows exponentially in time for the superposition initial state, the required bond dimension remains manageable (around 1024 at cycle 30 on an 81-qubit grid) compared to the exponential growth expected from classical simulation limits. This confirms that MPS simulations can accurately capture the dynamics of these systems, even when simulating complex, disorder-free states.

Tunable Initial States and Long-Time Dynamics

The study investigates a tunable family of initial states parameterized by an angle θ in Eq. (S12), which are "disorder-free for all values of a parameter θ.

Improvements for AI systems

As a fastidious and diligent researcher, I have analyzed this fascinating work on disorder-free localization (DFL) in lattice gauge theories (LGT) using quantum hardware. The core findings revolve around leveraging translationally invariant states as superpositions over all disorder configurations to achieve an efficient sampling of disordered systems, offering a polynomial speedup over traditional methods for many-body localization (MBL).

Here are the specific improvements and capabilities this research suggests for AI systems:

)Specific Improvements & Enhanced AI Capabilities)

(Polynomial Speedup in Disorder Sampling):
(Efficient Rare Event Simulation):
(Robust Characterization of Non-Ergodic Phases):
(Quantum Algorithm for Disordered Optimization):
(High-Fidelity Quantum Simulation of Complex Hamiltonians):
A quantum algorithm leveraging the superposition over all disorder configurations technique to sample quenched disorder realizations with a polynomial speedup in terms of the number of required quantum circuit executions, rather than the exponential sampling cost required by traditional methods (e.g., direct Monte Carlo or repeated Hamiltonian evolution).
The ability to simulate and characterize non-ergodic phases (like Many-Body Localization) in interacting systems more efficiently by utilizing translationally invariant initial states, allowing for the rapid identification of localization signatures without needing to average over an exponential number of disorder realizations. This enables faster diagnostics for complex material science problems involving disordered quantum materials.
The development of a robust framework for probing and distinguishing between different types of non-ergodic behavior (DFL vs. MBL) in lattice gauge theories, which can be mapped onto physical models relevant to condensed matter physics and topological phases. This allows AI/ML models trained on quantum simulation data to more accurately classify the underlying physical mechanism causing transport suppression or localization in complex many-body systems.
A novel phase estimation algorithm (based on Grover search) that can estimate disorder-averaged expectation values of Pauli observables with high accuracy in a polynomial number of applications, providing a significant advantage for searching for rare events—a critical step in understanding the stability and transition points between localized and thermal phases in quantum many-body systems.
A methodology for performing high-fidelity quantum simulations of complex, interacting Hamiltonians (like LGTs) by employing Matrix Product States (MPS) with optimized bond dimensions, allowing for the simulation of larger system sizes and longer time scales than currently feasible. This capability is crucial for testing new quantum algorithms and simulating more realistic physical models that involve strong interactions and disorder.

Abstract

Disorder-induced phenomena in quantum many-body systems pose significant challenges for analytical methods and numerical simulations at relevant time and system scales. To reduce the cost of disorder-sampling, we investigate quantum circuits initialized in states tunable to superpositions over all disorder configurations. In a translationally-invariant lattice gauge theory (LGT), these states can be interpreted as a superposition over gauge sectors. We observe localization in this LGT in the absence of disorder in one and two dimensions: perturbations fail to diffuse despite fully disorder-free evolution and initial states. However, Rényi entropy measurements reveal that superposition-prepared states fundamentally differ from those obtained by direct disorder sampling. Leveraging superposition, we propose an algorithm with a polynomial speedup in sampling disorder configurations, a longstanding challenge in many-body localization studies.