Mitigating disorder and optimizing topological indicators with vision-transformer-based neural networks in Majorana nanowires

Abstract

Disorder remains a major obstacle to realizing topological Majorana zero modes (MZMs) in superconductor-semiconductor nanowires, and we show how deep learning can be used to recover topological MZMs mitigating disorder even when the pre-mitigation situation manifests no apparent topology. The disorder potential, as well as the scattering invariant ($T_V$) normally used to classify a device as topologically non-trivial are not directly measurable experimentally. Additionally, the conventional signatures of MZMs have proved insufficient due to their being accidentally replicated by disorder-induced trivial states. Recent advances in machine learning provide a novel method to solve these problems, allowing the underlying topology, suppressed by disorder, to be recovered using effective mitigation procedures. In this work, we leverage a vision transformer neural network trained on conductance measurements along with a CMA-ES optimization framework to dynamically tune gate voltages mitigating disorder effects. Unlike prior efforts that relied on indirect cost functions, our method directly optimizes $T_V$ alongside additional local density of states-based topological indicators. Using a lightweight neural network variant, we demonstrate that even highly disordered nanowires initially lacking any topologically non-trivial regions can be transformed into robust topological devices.

arXiv comment14 figures, 10 pages, Updated published version with a new title enforced by the journal editors

References in corpus (34)

• Fault-tolerant quantum computation by anyons
• Unpaired Majorana fermions in quantum wires
• Signatures of Majorana fermions in hybrid superconductor-semiconductor nanowire devices
• Majorana Fermions and a Topological Phase Transition in Semiconductor-Superconductor Heterostructures
• Helical liquids and Majorana bound states in quantum wires
• Evidence of Majorana fermions in an Al - InAs nanowire topological superconductor
• A generic new platform for topological quantum computation using semiconductor heterostructures
• Observation of Majorana Fermions in a Nb-InSb Nanowire-Nb Hybrid Quantum Device
• Kwant: a software package for quantum transport
• Majorana Zero Modes and Topological Quantum Computation
• Realizing Majorana zero modes in superconductor-semiconductor heterostructures
• Superconductor-Nanowire Devices from Tunneling to the Multichannel Regime: Zero-Bias Oscillations and Magnetoconductance Crossover
• Anomalous modulation of a zero bias peak in a hybrid nanowire-superconductor device
• Zero-bias peaks in spin-orbit coupled superconducting wires with and without Majorana end-states
• Scaling of Majorana Zero-Bias Conductance Peaks
• Quantized conductance at the Majorana phase transition in a disordered superconducting wire
• Midgap edge states and pairing symmetry of quasi-one-dimensional organic superconductors
• Physical mechanisms for zero-bias conductance peaks in Majorana nanowires
• InAs-Al Hybrid Devices Passing the Topological Gap Protocol
• Robustness of Majorana fermions in 2D topological superconductors
• Non-Majorana states yield nearly quantized conductance in superconductor-semiconductor nanowire devices
• Disorder-induced zero-bias peaks in Majorana nanowires
• Estimating disorder and its adverse effects in semiconductor Majorana nanowires
• Three-terminal nonlocal conductance in Majorana nanowires: Distinguishing topological and trivial in realistic systems with disorder and inhomogeneous potential
• In search of Majorana
• Large zero bias peaks and dips in a four-terminal thin InAs-Al nanowire device
• Spectral properties, topological patches, and effective phase diagrams of finite disordered Majorana nanowires
• Density of states, transport, and topology in disordered Majorana nanowires
• Machine learning optimization of Majorana hybrid nanowires
• Neural network based deep learning analysis of semiconductor quantum dot qubits for automated control
• Disordered Majorana nanowires: Studying disorder without any disorder
• Machine Learning the Disorder Landscape of Majorana Nanowires
• Vision transformer based Deep Learning of Topological indicators in Majorana Nanowires
• Topological gap protocol based machine learning optimization of Majorana hybrid wires

Cited by in corpus (1)

• AI-enhanced tuning of quantum dot Hamiltonians toward Majorana modes