arXiv subjects

Derek Wang

Publications and source records attributed to Derek Wang.

4 recordsLinked to original sources

arXiv2026

Hedgementation = Hedgerow Segmentation: A Remote Sensing Benchmark

We propose Hedgementation: a new benchmark to evaluate machine learning models for hedgerow mapping from remote sensing data at country scale and 10m$^2$ spatial resolution. We combine and harmonize multiple remote sensing data products and ground truth labels sourced from a hedgerow inventory in France. We measure the ability of three baseline models to generalize across spatial distance, and across climatic zones, a more explicitly challenging task. Our benchmark tests both supervised and self-supervised learning approaches for remote sensing, applied to tracking fine-scale features of high agricultural importance. The code to reproduce the benchmark and baselines results is available at https://github.com/hedgementation/hedgementation.

cs.CV↗

arXiv2025

Pauli Network Circuit Synthesis with Reinforcement Learning

We introduce a Reinforcement Learning (RL)-based method for re-synthesis of quantum circuits containing arbitrary Pauli rotations alongside Clifford operations. By collapsing each sub-block to a compact representation and then synthesizing it step-by-step through a learned heuristic, we obtain circuits that are both shorter and compliant with hardware connectivity constraints. We find that the method is fast enough and good enough to work as an optimization procedure: in direct comparisons on 6-qubit random Pauli Networks against state-of-the-art heuristic methods, our RL approach yields over 2x reduction in two-qubit gate count, while executing in under 10 milliseconds per circuit. We further integrate the method into a collect-and-re-synthesize pipeline, applied as a Qiskit transpiler pass, where we observe average improvements of 20% in two-qubit gate count and depth, reaching up to 60% for many instances, across the Benchpress benchmark. These results highlight the potential of RL-driven synthesis to significantly improve circuit quality in realistic, large-scale quantum transpilation workloads.

quant-ph↗

arXiv2018

Scalable, "Dip-and-dry" Fabrication of a Wide-Angle Plasmonic Selective Absorber for High-efficiency Solar-Thermal Energy Conversion

A galvanic displacement reaction-based, room-temperature "dip-and-dry" technique is demonstrated for fabricating selectively solar-absorbing plasmonic nanostructure-coated foils (PNFs). The technique, which allows for facile tuning of the PNFs' spectral reflectance to suit different radiative and thermal environments, yields PNFs which exhibit excellent, wide-angle solar absorptance (0.96 at 15°, to 0.97 at 35°, to 0.79 at 80°) and low hemispherical thermal emittance (0.10) without the aid of antireflection coatings. The thermal emittance is on par with those of notable selective solar absorbers (SSAs) in the literature, while the wide-angle solar absorptance surpasses those of previously reported SSAs with comparable optical selectivities. In addition, the PNFs show promising mechanical and thermal stabilities at temperatures of up to 200°C. Along with the performance of the PNFs, the simplicity, inexpensiveness and environment-friendliness of the "dip-and-dry" technique makes it an appealing alternative to current methods for fabricating selective solar absorbers.

physics.app-ph↗

arXiv2018

Defending against Adversarial Attack towards Deep Neural Networks via Collaborative Multi-task Training

Deep neural networks (DNNs) are known to be vulnerable to adversarial examples which contain human-imperceptible perturbations. A series of defending methods, either proactive defence or reactive defence, have been proposed in the recent years. However, most of the methods can only handle specific attacks. For example, proactive defending methods are invalid against grey-box or white-box attacks, while reactive defending methods are challenged by low-distortion adversarial examples or transferring adversarial examples. This becomes a critical problem since a defender usually does not have the type of the attack as a priori knowledge. Moreover, existing two-pronged defences (e.g., MagNet), which take advantages of both proactive and reactive methods, have been reported as broken under transferring attacks. To address this problem, this paper proposed a novel defensive framework based on collaborative multi-task training, aiming at providing defence for different types of attacks. The proposed defence first encodes training labels into label pairs and counters black-box attacks leveraging adversarial training supervised by the encoded label pairs. The defence further constructs a detector to identify and reject high-confidence adversarial examples that bypass the black-box defence. In addition, the proposed collaborative architecture can prevent adversaries from finding valid adversarial examples when the defence strategy is exposed. In the experiments, we evaluated our defence against four state-of-the-art attacks on $MNIST$ and $CIFAR10$ datasets. The results showed that our defending method achieved up to $96.3\%$ classification accuracy on black-box adversarial examples, and detected up to $98.7\%$ of the high confidence adversarial examples. It only decreased the model accuracy on benign example classification by $2.1\%$ for the $CIFAR10$ dataset.

cs.LG↗