arXiv:2505.02979v2 Announce Type: replace-cross 
Abstract: The Neural Physics approach is used to determine the parameters of a simple land-surface model using PyTorch's backpropagation engine to carry out the optimisation. In order to test the inverse model, a synthetic dataset is created by running the model in forward mode with known parameter values to create soil temperature time series that can be used as observations for the inverse model. We show that it is not possible to obtain a reliable parameter estimation using a time series of soil temperature observed at a single depth. Using measurements at two depths, reliable parameter estimates can be obtained although it is not possible to differentiate between latent and sensible heat fluxes. We apply the inverse model to urban flux tower data in Phoenix, United States, and show that the thermal conductivity, volumetric heat capacity and the combined sensible-latent heat transfer coefficient can be reliably estimated using an observed value for the effective surface albedo. The resulting model accurately predicts the outgoing longwave radiation, conductive soil fluxes and the combined sensible-latent heat fluxes.

تم استخدام نهج الفيزياء العصبية لتقدير معلمات نموذج بسيط لسطح الأرض، باستخدام محرك الانتشار العكسي من PyTorch لإجراء التحسين. تم إنشاء مجموعة بيانات اصطناعية لاختبار النموذج العكسي، مما يكشف أن التقدير الموثوق للمعلمات يتطلب قياسات لدرجة حرارة التربة على عمقين بدلاً من عمق واحد. تم تطبيق النموذج على بيانات أبراج التدفق الحضرية في فينيكس، مما سمح بتقدير الخصائص الحرارية الرئيسية.

Se ha utilizado el enfoque de Física Neural para estimar parámetros de un modelo simple de superficie terrestre, empleando el motor de retropropagación de PyTorch para la optimización. Se generó un conjunto de datos sintético para probar el modelo inverso, revelando que la estimación confiable de parámetros requiere mediciones de temperatura del suelo a dos profundidades en lugar de una. El modelo se aplicó a datos de torres de flujo urbanas en Phoenix, permitiendo la estimación de propiedades térmicas clave.

L'approche Neural Physics a été utilisée pour estimer les paramètres d'un modèle simple de surface terrestre, en utilisant le moteur de rétropropagation de PyTorch pour l'optimisation. Un ensemble de données synthétiques a été généré pour tester le modèle inverse, révélant que l'estimation fiable des paramètres nécessite des mesures de température du sol à deux profondeurs plutôt qu'à une seule. Le modèle a été appliqué aux données de tours de flux urbains à Phoenix, permettant d'estimer des propriétés thermiques clés.

The Neural Physics approach has been utilized to estimate parameters for a simple land-surface model, employing PyTorch's backpropagation engine for optimization. A synthetic dataset was generated to test the inverse model, revealing that reliable parameter estimation requires soil temperature measurements at two depths rather than one. The model was applied to urban flux tower data in Phoenix, allowing for the estimation of key thermal properties.

Parameter estimation for land-surface models using Neural Physics

arXiv:2512.07860v1 Announce Type: cross 
Abstract: This paper investigates an optimal integration of deep learning with financial models for robust asset price forecasting. Specifically, we developed a hybrid framework combining a Long Short-Term Memory (LSTM) network with the Merton-L\'evy jump-diffusion model. To optimise this framework, we employed the Grey Wolf Optimizer (GWO) for the LSTM hyperparameter tuning, and we explored three calibration methods for the Merton-Levy model parameters: Artificial Neural Networks (ANNs), the Marine Predators Algorithm (MPA), and the PyTorch-based TorchSDE library. To evaluate the predictive performance of our hybrid model, we compared it against several benchmark models, including a standard LSTM and an LSTM combined with the Fractional Heston model. This evaluation used three real-world financial datasets: Brent oil prices, the STOXX 600 index, and the IT40 index. Performance was assessed using standard metrics, including Mean Squared Error (MSE), Mean Absolute Error(MAE), Mean Squared Percentage Error (MSPE), and the coefficient of determination (R2). Our experimental results demonstrate that the hybrid model, combining a GWO-optimized LSTM network with the Levy-Merton Jump-Diffusion model calibrated using an ANN, outperformed the base LSTM model and all other models developed in this study.

قدمت دراسة حديثة إطارًا هجينًا يدمج الشبكات ذات الذاكرة طويلة المدى (LSTM) مع نموذج مارتون-ليفي للقفز والانحراف في التنبؤ المالي. تستخدم هذه الطريقة مُحسّن الذئب الرمادي لضبط المعلمات وتقيّم أدائها التنبؤي مقارنةً بالنماذج المرجعية باستخدام مجموعات بيانات حقيقية، بما في ذلك أسعار النفط برنت ومؤشر STOXX 600.

Un estudio reciente ha presentado un marco híbrido que integra redes de memoria a largo plazo (LSTM) con el modelo de salto-difusión de Merton-Lévy para la previsión financiera. Este enfoque utiliza el Optimizador Grey Wolf para el ajuste de hiperparámetros y evalúa su rendimiento predictivo en comparación con modelos de referencia utilizando conjuntos de datos del mundo real, incluidos los precios del petróleo Brent y el índice STOXX 600.

Une étude récente a introduit un cadre hybride qui intègre les réseaux de mémoire à long terme (LSTM) avec le modèle de saut-diffusion de Merton-Lévy pour la prévision financière. Cette approche utilise l'optimiseur Grey Wolf pour le réglage des hyperparamètres et évalue sa performance prédictive par rapport à des modèles de référence en utilisant des ensembles de données réelles, y compris les prix du pétrole Brent et l'indice STOXX 600.

A recent study has introduced a hybrid framework that integrates Long Short-Term Memory (LSTM) networks with the Merton-Lévy jump-diffusion model for financial forecasting. This approach utilizes the Grey Wolf Optimizer for hyperparameter tuning and evaluates its predictive performance against benchmark models using real-world datasets, including Brent oil prices and the STOXX 600 index.

Integrating LSTM Networks with Neural Levy Processes for Financial Forecasting

<img width="800" height="450" src="https://analyticsindiamag.com/wp-content/uploads/2025/12/trump-jensen-huang-nvidia-china-US-1300x731.jpg" class="webfeedsFeaturedVisual wp-post-image" alt="" style="display: block; margin: auto; margin-bottom: 5px;max-width: 100%;" link_thumbnail="" decoding="async" loading="lazy" srcset="https://analyticsindiamag.com/wp-content/uploads/2025/12/trump-jensen-huang-nvidia-china-US-1300x731.jpg 1300w, https://analyticsindiamag.com/wp-content/uploads/2025/12/trump-jensen-huang-nvidia-china-US-768x432.jpg 768w, https://analyticsindiamag.com/wp-content/uploads/2025/12/trump-jensen-huang-nvidia-china-US-1536x864.jpg 1536w, https://analyticsindiamag.com/wp-content/uploads/2025/12/trump-jensen-huang-nvidia-china-US-2048x1152.jpg 2048w" sizes="auto, (max-width: 800px) 100vw, 800px" /><p>The 25% fee will be charged when the chips are shipped from Taiwan to the United States for inspection, before onward export.</p>
<p>The post <a href="https://analyticsindiamag.com/ai-news-updates/nvidia-gains-us-approval-to-export-h200-ai-chips-to-china/">NVIDIA Gains US Approval to Export H200 AI Chips to China</a> appeared first on <a href="https://analyticsindiamag.com">Analytics India Magazine</a>.</p>

وافقت إدارة ترامب على تصدير شركة NVIDIA لشرائح الذكاء الاصطناعي H200 إلى الصين، بعد مفاوضات مكثفة. تتضمن هذه القرار فرض رسوم بنسبة 25% على المبيعات، والتي سيتم تحصيلها عند شحن الشرائح من تايوان إلى الولايات المتحدة للتفتيش قبل التصدير الإضافي.

La administración Trump ha otorgado la aprobación a NVIDIA para exportar sus chips de inteligencia artificial H200 a China, tras extensas negociaciones. Esta decisión incluye una estipulación de un 25% de tarifa sobre las ventas, que se cobrará cuando los chips sean enviados desde Taiwán a Estados Unidos para inspección antes de la exportación adicional.

L'administration Trump a accordé l'approbation à NVIDIA pour exporter ses puces d'intelligence artificielle H200 vers la Chine, suite à des négociations approfondies. Cette décision inclut une stipulation d'une taxe de 25 % sur les ventes, qui sera appliquée lorsque les puces seront expédiées de Taïwan vers les États-Unis pour inspection avant une nouvelle exportation.

The Trump administration has granted approval for NVIDIA to export its H200 artificial intelligence chips to China, following extensive negotiations. This decision includes a stipulation of a 25% fee on sales, which will be charged when the chips are shipped from Taiwan to the U.S. for inspection before further export.

Parameter estimation for land-surface models using Neural Physics

Was this article worth reading? Share it

LucidQuery AI

Airparser

Hypertune