Hybrid Classical–Quantum Transfer Learning Using Variational Quantum Layers for Low-Data Regimes

Fuente: Zenodo
Saved in:
Bibliographic Details
Main Authors: Munishamaiah Krishna, Munishamaiah Krishna, Munishamaiah Krishna
Format: Recurso digital
Published: Zenodo 2026
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866901780224278528
author Munishamaiah Krishna, Munishamaiah Krishna
Munishamaiah Krishna
author_facet Munishamaiah Krishna, Munishamaiah Krishna
Munishamaiah Krishna
contents <h2><strong>Supplementary Data Description</strong></h2> <p>This supplementary dataset provides the complete experimental results underlying Figures 4–9 of the manuscript. Each CSV file corresponds to one figure and reports mean performance values (and standard deviations where applicable), computed over multiple independent runs with different random seeds.</p> <h3><strong>Fig4_Low_Data_Performance.csv</strong></h3> <p>Contains classification accuracy (mean ± standard deviation) for Linear, MLP, and VQL heads across varying labeled target data fractions (1%, 5%, 10%, 20%). This dataset supports the low-data performance analysis presented in Fig. 4.</p> <h3><strong>Fig5_Parameter_Matched.csv</strong></h3> <p>Reports accuracy statistics for parameter-matched adaptation heads, including Linear Probe, MLP Adapter, Bottleneck Adapter, and the proposed VQL Head. This file enables fair comparison of adaptation strategies under comparable trainable parameter budgets (Fig. 5).</p> <h3><strong>Fig6_Domain_Shift.csv</strong></h3> <p>Provides source-domain accuracy, target-domain accuracy, and the resulting accuracy drop for each model under domain shift conditions. These results quantify robustness to distributional changes, as shown in Fig. 6.</p> <h3><strong>Fig7_Ablation.csv</strong></h3> <p>Contains ablation results evaluating the effect of quantum circuit depth and number of qubits on classification accuracy. This dataset supports the architectural analysis in Fig. 7.</p> <h3><strong>Fig8_Noise_Aware.csv</strong></h3> <p>Includes classification accuracy values under increasing quantum noise probabilities for all adaptation heads, enabling noise robustness evaluation corresponding to Fig. 8.</p> <h3><strong>Fig9_Convergence.csv</strong></h3> <p>Reports training accuracy as a function of training epochs for Linear, MLP, and VQL heads, illustrating convergence behavior and optimization stability (Fig. 9).</p> <p><strong>Reproducibility Note</strong></p> <p>All values are reported as mean performance over multiple independent runs, ensuring statistical reliability and reproducibility of the experimental findings.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_18159176
institution Zenodo
language
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle Hybrid Classical–Quantum Transfer Learning Using Variational Quantum Layers for Low-Data Regimes
Munishamaiah Krishna, Munishamaiah Krishna
Munishamaiah Krishna
<h2><strong>Supplementary Data Description</strong></h2> <p>This supplementary dataset provides the complete experimental results underlying Figures 4–9 of the manuscript. Each CSV file corresponds to one figure and reports mean performance values (and standard deviations where applicable), computed over multiple independent runs with different random seeds.</p> <h3><strong>Fig4_Low_Data_Performance.csv</strong></h3> <p>Contains classification accuracy (mean ± standard deviation) for Linear, MLP, and VQL heads across varying labeled target data fractions (1%, 5%, 10%, 20%). This dataset supports the low-data performance analysis presented in Fig. 4.</p> <h3><strong>Fig5_Parameter_Matched.csv</strong></h3> <p>Reports accuracy statistics for parameter-matched adaptation heads, including Linear Probe, MLP Adapter, Bottleneck Adapter, and the proposed VQL Head. This file enables fair comparison of adaptation strategies under comparable trainable parameter budgets (Fig. 5).</p> <h3><strong>Fig6_Domain_Shift.csv</strong></h3> <p>Provides source-domain accuracy, target-domain accuracy, and the resulting accuracy drop for each model under domain shift conditions. These results quantify robustness to distributional changes, as shown in Fig. 6.</p> <h3><strong>Fig7_Ablation.csv</strong></h3> <p>Contains ablation results evaluating the effect of quantum circuit depth and number of qubits on classification accuracy. This dataset supports the architectural analysis in Fig. 7.</p> <h3><strong>Fig8_Noise_Aware.csv</strong></h3> <p>Includes classification accuracy values under increasing quantum noise probabilities for all adaptation heads, enabling noise robustness evaluation corresponding to Fig. 8.</p> <h3><strong>Fig9_Convergence.csv</strong></h3> <p>Reports training accuracy as a function of training epochs for Linear, MLP, and VQL heads, illustrating convergence behavior and optimization stability (Fig. 9).</p> <p><strong>Reproducibility Note</strong></p> <p>All values are reported as mean performance over multiple independent runs, ensuring statistical reliability and reproducibility of the experimental findings.</p>
title Hybrid Classical–Quantum Transfer Learning Using Variational Quantum Layers for Low-Data Regimes
url https://doi.org/10.5281/zenodo.18159176