LIVE PULSE
4.0 Anthropic CEO Amodei calls for slower AI development and shared safety rules — 11 src2.2 Agility Robotics unveils Digit 5 humanoid for warehouses and factories — 2 src2.0 Apple ships rebuilt Siri with Google Gemini, but not in the EU — 2 src1.8 Siri AI in macOS 27 Golden Gate: FAQ, Germany availability, privacy questions — 2 src1.4 Sam Altman says OpenAI will not go public in 2026, citing AI safety concerns — 5 src1.1 OpenAI contractors review real ChatGPT conversations to rate responses, report says — 2 src1.1 Anthropic data retention policy prompts firms to limit Claude use for sensitive work — 1 src1.1 Study examines issue bias in LLMs used as writing assistants before Swedish 2026 election — 1 src1.1 Study Audits Misalignment in Multi-Modal World Models — 1 src1.1 Retrieval-Grounded Reasoning Approach Proposed for Universal Multimodal Embeddings — 1 src4.0 Anthropic CEO Amodei calls for slower AI development and shared safety rules — 11 src2.2 Agility Robotics unveils Digit 5 humanoid for warehouses and factories — 2 src2.0 Apple ships rebuilt Siri with Google Gemini, but not in the EU — 2 src1.8 Siri AI in macOS 27 Golden Gate: FAQ, Germany availability, privacy questions — 2 src1.4 Sam Altman says OpenAI will not go public in 2026, citing AI safety concerns — 5 src1.1 OpenAI contractors review real ChatGPT conversations to rate responses, report says — 2 src1.1 Anthropic data retention policy prompts firms to limit Claude use for sensitive work — 1 src1.1 Study examines issue bias in LLMs used as writing assistants before Swedish 2026 election — 1 src1.1 Study Audits Misalignment in Multi-Modal World Models — 1 src1.1 Retrieval-Grounded Reasoning Approach Proposed for Universal Multimodal Embeddings — 1 src
HEATPULSEAI MAGAZINES
FLIP · FOLLOW · SAVE
Log inSign up

surrogate modeling

topic6 events
papersTODAY 04:00 UTC

Active Learning with Bayesian Multi-Fidelity Laplace Neural Operators for Parametric PDEs

This arXiv paper proposes a surrogate modeling approach that combines Laplace neural operators with multi-fidelity Bayesian active learning for oscillatory parametric partial differential equations. High-fidelity simulation data is costly to generate, so the method uses Bayesian uncertainty estimates to decide which fidelity levels and parameter points to sample next. The authors frame the work around engineering uses such as design optimization and digital twins, where fast, repeated predictions are needed.

research/active-learning/pde/surrogate-models
papersTODAY 04:00 UTC

Multifidelity TDNN and Physics-Informed Residual Learning for Railway Bogie Prediction

A revised arXiv preprint proposes a hybrid approach for predicting railway bogie responses, combining a time-delay neural network with physics-informed residual learning across simulations of differing fidelity. The method targets operating conditions that are impractical to test exhaustively, using agreement with representative measurements as validation evidence. It sits within ongoing work on surrogate modeling for engineering simulation.

railways/simulation/physics-informed/neural-networks
papersTODAY 04:00 UTC

Bayesian Optimisation Method Combines Expert Gaussian Processes With Calibrated Uncertainty

A new arXiv preprint proposes using a product-of-experts Gaussian process as the surrogate model in Bayesian optimisation, rather than a single global GP. The authors address the cubic scaling cost of standard GP regression with training set size, which restricts its use on larger datasets, and add a calibration step for uncertainty estimates. The work falls in the machine learning methodology category and has not yet been peer reviewed.

bayesian-optimisation/gaussian-processes/uncertainty/arxiv
papersSEP 10 04:00 UTC

Graph Neural Operator Surrogate Predicts Stress Tensor Fields in Concrete Penetration

Researchers have developed a graph neural operator that estimates full mid-plane stress tensor fields in concrete during projectile penetration, connecting the material's mesoscale structure to complete field-level predictions. The surrogate was trained on data from a detailed aggregate-resolved LS-DYNA simulation and is designed to generalize across different impact velocities, offering a faster stand-in for expensive finite-element computations.

graph-neural-networks/surrogate-models/simulation/materials-science
papersSEP 10 04:00 UTC

Convolutional autoencoder and neural ODE framework for transient counterflow flame modeling

Researchers propose a reduced-order modeling framework that combines a convolutional autoencoder with a neural ordinary differential equation to serve as a surrogate for simulating transient two-dimensional counterflow flames. The approach extends autoencoder–neural ODE techniques, previously applied to homogeneous reactive systems, to spatially resolved combustion problems. Such surrogates can cut the computational cost of modeling reactive flows.

scientific-ml/neural-ode/combustion/surrogate-models
papersSEP 10 04:00 UTC

Researchers build physics-informed surrogate model for Mars' nightside thermosphere

A new arXiv paper introduces a multi-task surrogate model that combines physical constraints with machine learning to simulate the Martian nightside thermosphere. The problem is difficult because direct measurements are sparse and transport, magnetic, and seasonal effects interact strongly, so purely data-driven approaches can produce unphysical outputs such as reversed density trends. Embedding physics into the training process aims to keep the model's predictions consistent with known atmospheric behavior.

space/physics-informed/machine-learning/research