The paper introduces SMART, a lightweight recalibration technique that adjusts logits based on the margin between the top two logits, called the logit gap. It uses a soft-binned Expected Calibration Error objective to balance bias and variance, enabling stable updates even with limited calibration data. Experiments across various datasets and architectures show SMART achieves state‑of‑the‑art calibration with fewer parameters than existing methods.
By Haolan Guo, Linwei Tao, Haoyang Luo, Minjing Dong, Chang Xu
arXiv:2605. 02524v2 Announce Type: replace Abstract: Accurate reconstruction of greenhouse climate variables from sparse sensor measurements is essential for intelligent environmental monitoring, automated climate control, and precision agriculture.
By Sani Biswas, Khursheed J. Ansari, Md. Nasim Akhtar
The paper investigates how temperature affects analog deep neural network (DNN) inference, focusing on both stochastic and systematic non‑idealities in analog hardware. Experiments show that temperature‑induced performance loss is mainly driven by systematic errors rather than random noise. The study evaluates various mitigation techniques, finding that noise‑aware training and temperature‑aware calibration—especially hardware‑in‑the‑loop training—best preserve inference accuracy across different thermal conditions.
By Niklas Summ, Xiao Wang, Hendrik Borras, Bernhard Klein, Holger Fr\"oning
arXiv:2602. 02547v2 Announce Type: replace-cross Abstract: Physics-Informed Neural Networks (PINNs) are effective methods for solving inverse problems and discovering governing equations from observational data.
By Hankyeol Kim, Pilsung Kang
arXiv:2510. 15968v2 Announce Type: replace-cross Abstract: Thermal management in 3D ICs is increasingly challenging due to higher power densities.
By Zhen Huang, Hong Wang, Wenkai Yang, Muxi Tang, Depeng Xie, Ting-Jung Lin, Yu Zhang, Wei W. Xing, Lei He
arXiv:2608. 10941v1 Announce Type: new Abstract: Industrial time-series signals, such as turbine temperature and rotational speed in aero-engines, are essential for monitoring the health and operational status of complex dynamical systems.
By Haiteng Wang, Yunfei Zhu, Tao Wang, Yikang Li, Jiabao Dong, Xiaoge Zhang, Lei Ren
The paper introduces Langevin simulated bifurcation (LSB), a fast, parallel Boltzmann sampler that matches the accuracy of sequential MCMC methods. It also proposes conditional expectation matching (CEM), an efficient technique for estimating the effective temperature of samples from energy‑based models with conditional independence. Building on these, the authors develop sampler adaptive learning (SAL), which adjusts the model temperature to align with the distribution produced by LSB, enabling efficient training of semi‑restricted Boltzmann machines (SRBMs) and outperforming conventional methods on synthetic spin‑glass datasets.
By Kentaro Kubo, Hayato Goto
The paper introduces Shallow Recurrent Decoder (SHRED) networks as a data‑driven method for accurate state estimation in engineering systems, specifically applied to the TRIGA Mark II research reactor. SHRED maps sparse sensor measurements to the full state space, handling noisy data and requiring minimal training time. The study demonstrates SHRED’s performance using both synthetic CFD data and experimental temperature recordings, achieving low reconstruction errors and showcasing its potential for real‑time monitoring and digital twin development.
By Stefano Riva, Carolina Introini, Jos\`e Nathan Kutz, Antonio Cammi
arXiv:2609.15439v1 Announce Type: cross
Abstract: Generative thermodynamic computers turn thermal noise into structured data through Langevin dynamics. We train these systems with a local update at e...
By Huilin Wang, Weibing Deng
arXiv:2606. 11605v1 Announce Type: cross Abstract: Predicting process-property relationships in manufacturing is often challenged by high experimental costs and the limited interpretability of complex 'black-box' models.
By Ge Song, Kiarash Naghavi Khanghah, Anandkumar Patel, Rajiv Malhotra, Hongyi Xu
The paper proposes a physics-informed neural network (PINN) architecture that uses the complex exponential function as an activation, producing sine and cosine outputs. Compared to standard sine-activated multilayer perceptrons, this design yields markedly better temperature reconstructions from sparse velocity data in cubic Rayleigh-Bénard convection, while keeping per‑step computational cost similar. The authors attribute the gains to the network’s ability to propagate both sine and cosine components, allowing each neuron in the next layer to adjust the phase of the latent periodic signals.
By Michael Mommert, Marie-Christine Volk, Christian Bauer
arXiv:2607. 16183v1 Announce Type: new Abstract: To address the escalating energy and latency demands of machine-learning workloads, we introduce a blueprint for an energy-efficient and fast thermodynamic computing stack that leverages stochastic analog processes in physical hardware.
By Owen Lockwood, J\'er\'emy B\'ejanin, Joost Bus, Christopher Chamberland, Patrick Huembeli, Frank Sch\"afer, Guillaume Verdon