The paper introduces a reinforcement‑learning framework that automatically discovers compact parametrized quantum circuits for modeling power GaN HEMTs and logic nanowire FETs. Using a graph neural network policy trained with proximal policy optimization, the method optimizes circuit architectures based on leave‑one‑group‑out cross‑validation error, achieving the lowest mean absolute error across 11 targets compared to six classical baselines. The results show significant reductions in error and variability for key device metrics (Ioff, VTH, SS) on both HEMT and NWFET datasets, demonstrating the viability of RL‑selected quantum circuits as compact, physically consistent surrogates without explicit physical constraints.
By Rushat Rai, Yun-Yuan Wang, Autsada Kakaen, Pei-Jie Chang, Doan Viet Nguyen, Yuan-Chieh Chiu, Doldet Tantraviwat, Niall Tumilty, Simon See, Wen-Jay Lee, Tai-Yue Li, Nan-Yow Chen, Tian-Li Wu
arXiv:2607. 00961v1 Announce Type: cross Abstract: Realizing quantum neural networks (QNNs) in industry requires knowing which quantum computing paradigm suits which task.
By Yeonhong Kim, Jonghyeok Im, Monu Nath Baitha, Kyoungsik Kim
arXiv:2605. 27923v2 Announce Type: replace-cross Abstract: The rapid growth of computer vision and increasingly complex image recognition tasks has exposed fundamental computational limitations of classical machine learning models, motivating the exploration of quantum computing as an emerging new paradigm.
By Sudip Vhaduri, Ryan Gammon, Sayanton Dibbo
The paper systematically compares four classical machine learning models—SVM, ANN, CNN, and LSTM—with their quantum equivalents—QSVM, QNN, QCNN, and QLSTM—on simulated proton‑proton collision data from CERN Open Data. Classical models, especially CNN and LSTM, slightly outperform the quantum models under current hardware and dataset constraints, but quantum models achieve comparable accuracy with far fewer trainable parameters; for example, the QCNN matches a deep classical CNN using only four qubits and a depth‑three circuit. The study also shows that the regression task is non‑trivial for shallow polynomial fits, underscoring the relevance of the architectural comparison.
whyItMatters":"The work provides a realistic benchmark of classical versus quantum machine learning performance on high‑energy physics data, highlighting parameter‑efficiency advantages of quantum models for near‑term devices."
By Tariq Mahmood, Zain ul Abidin, Itzel Luviano Soto, Alfredo Raya
arXiv:2510. 03389v2 Announce Type: replace-cross Abstract: Current quantum computers require algorithms that use limited resources economically.
By Jonas J\"ager, Philipp Els\"asser, Elham Torabian
arXiv:2607. 14293v1 Announce Type: cross Abstract: Ionizing radiation from cosmic rays and gammas can induce discontinuous jumps in the environmental charge of superconducting qubits (charge jumps), causing correlated errors that challenge fault-tolerant quantum computing while simultaneously providing a detection signature for quantum sensing applications.
By Daniel Gaytan-Villarreal, Peter Meiring, Daniel Baxter, Daniel Bowring, Grace Bratrud, Matteo Cremonesi, Giuseppe Di Guglielmo, Grace Wagner, Bowen Xiao
arXiv:2604. 06135v2 Announce Type: replace-cross Abstract: Efficient data loading remains a bottleneck for near-term quantum machine learning.
By Basil Kyriacou, Viktoria Patapovich, Maniraman Periyasamy, Alexey Melnikov
arXiv:2607. 10388v1 Announce Type: cross Abstract: Artificial intelligence (AI) is transforming electron microscopy by enabling quantitative analysis of increasingly large and complex datasets for nanoparticle characterization.
By Evropi Toulkeridou, Jiafei Li, Leonardo Lari, Panagiotis Grammatikopoulos
arXiv:2412. 09486v2 Announce Type: replace-cross Abstract: The literature reflects a mutually beneficial relationship between machine learning and quantum computing, where progress in one field frequently drives improvements in the other.
By Leandro C. Souza, Bruno C. Guingo, Gilson Giraldi, Renato Portugal
TetrisCNN is a convolutional neural network that uses parallel branches of differently shaped filters to learn sparse, interpretable latent representations directly in terms of spin correlators. Applied to experimental snapshots from two-dimensional Ising and XY quantum simulators measured in multiple bases, the network detects phase transitions and crossovers while expressing its decision boundaries as symbolic formulas built from experimentally measurable spin correlators. This approach bridges the gap between black‑box neural network methods and physically interpretable models, enabling automated detection of phases of matter from realistic, noisy experimental data.
By Kacper Cybi\'nski, Bj\"orn van Zwol, James Enouen, Guillaume Bornet, Thierry Lahaye, Antoine Browaeys, Antoine Georges, Anna Dawid
TetrisCNN is a convolutional neural network that uses parallel branches of differently shaped filters to learn sparse, interpretable latent representations directly from spin correlators. Applied to experimental snapshots of two-dimensional Ising and XY quantum simulators, it detects phase transitions and crossovers while expressing its decision boundaries as symbolic formulas built from measurable spin correlators. This approach bridges the gap between black‑box neural networks and physically interpretable models, enabling automated discovery of new phases of matter from realistic, noisy experimental data.
arXiv:2601. 02149v4 Announce Type: replace-cross Abstract: We propose a neural network-based model capable of learning the broad landscape of working regimes in quantum dot simulators, and using this knowledge to autotune these devices - based on transport measurements - toward obtaining Majorana modes in the structure.
By Mateusz Krawczyk, Jaros{\l}aw Paw{\l}owski