arXiv subjects

Yen Jui Chang

Publications and source records attributed to Yen Jui Chang.

3 recordsLinked to original sources

arXiv2026

Photonic Quantum-Enhanced Knowledge Distillation

Photonic quantum processors naturally produce intrinsically stochastic measurement outcomes, offering a hardware-native source of structured randomness that can be exploited during machine-learning training. Here we introduce Photonic Quantum-Enhanced Knowledge Distillation (PQKD), a hybrid quantum photonic--classical framework in which a programmable photonic circuit generates a compact conditioning signal that constrains and guides a parameter-efficient student network during distillation from a high-capacity teacher. PQKD replaces fully trainable convolutional kernels with dictionary convolutions: each layer learns only a small set of shared spatial basis filters, while sample-dependent channel-mixing weights are derived from shot-limited photonic features and mapped through a fixed linear transform. Training alternates between standard gradient-based optimisation of the student and sampling-robust, gradient-free updates of photonic parameters, avoiding differentiation through photonic hardware. Across MNIST, Fashion-MNIST and CIFAR-10, PQKD traces a controllable compression--accuracy frontier, remaining close to teacher performance on simpler benchmarks under aggressive convolutional compression. Performance degrades predictably with finite sampling, consistent with shot-noise scaling, and exponential moving-average feature smoothing suppresses high-frequency shot-noise fluctuations, extending the practical operating regime at moderate shot budgets.

quant-ph↗

arXiv2026

QuIC: A Quantum-Inspired Interaction Classifier for Revitalizing Shallow CNNs in Fine-Grained Recognition

Deploying deep learning models for Fine-Grained Visual Classification (FGVC) on resource-constrained edge devices remains a significant challenge. While deep architectures achieve high accuracy on benchmarks like CUB-200-2011, their computational cost is often prohibitive. Conversely, shallow networks (e.g., AlexNet, VGG) offer efficiency but fail to distinguish visually similar sub-categories. This is because standard Global Average Pooling (GAP) heads capture only first-order statistics, missing the subtle high-order feature interactions required for FGVC. While Bilinear CNNs address this, they suffer from high feature dimensionality and instability during training. To bridge this gap, we propose the Quantum-inspired Interaction Classifier (QuIC). Drawing inspiration from quantum mechanics, QuIC models feature channels as interacting quantum states and captures second-order feature covariance via a learnable observable operator. Designed as a lightweight, plug-and-play module, QuIC supports stable, single-stage end-to-end training without exploding feature dimensions. Experimental results demonstrate that QuIC significantly revitalizes shallow backbones: it boosts the Top-1 accuracy of VGG16 by nearly 20% and outperforms state-of-the-art attention mechanisms (SE-Block) on ResNet18. Qualitative analysis, including t-SNE visualization, further confirms that QuIC resolves ambiguous cases by explicitly attending to fine-grained discriminative features and enforcing compact intra-class clustering.

cs.CV↗

arXiv2025

Quantum Stochastic Walks for Portfolio Optimization: Theory and Implementation on Financial Networks

Financial markets are noisy yet contain a latent graph-theoretic structure that can be exploited for superior risk-adjusted returns. We propose a quantum stochastic walk (QSW) optimizer that embeds assets in a weighted graph: nodes represent securities while edges encode the return-covariance kernel. Portfolio weights are derived from the walk's stationary distribution. Three empirical studies support the approach. (i) For the top 100 S\&P 500 constituents over 2016-2024, six scenario portfolios calibrated on 1- and 2-year windows lift the out-of-sample Sharpe ratio by up to 27\% while cutting annual turnover from 480\% (mean-variance) to 2-90%. (ii) A $5^{4}=625$-point grid search identifies a robust sweet spot, $α,λ\lesssim0.5$ and $ω\in[0.2,0.4]$, that delivers Sharpe $\approx0.97$ at $\le 5\%$ turnover and Herfindahl-Hirschman index $\sim0.01$. (iii) Repeating the full grid on 50 random 100-stock subsets of the S\&P 500 adds 31\,350 back-tests: the best-per-draw QSW beats re-optimised mean-variance on Sharpe in 54\% of cases and always wins on trading efficiency, with median turnover 36\% versus 351\%. Overall, QSW raises the annualized Sharpe ratio by 15\% and cuts turnover by 90\% relative to classical optimisation, all while respecting the UCITS 5/10/40 rule. These results show that hybrid quantum-classical dynamics can uncover non-linear dependencies overlooked by quadratic models and offer a practical, low-cost weighting engine for themed ETFs and other systematic mandates.

q-fin.PM↗