A new deep learning framework called DeepGATE reads analog gas meter dials from camera images in real time, improving measurement...

A new stem-encoder-decoder neural network runs crowd counting at up to 381.7 frames per second on desktop GPUs and 71.9...

A new systematic review of 121 studies finds that AI models reading chest X-rays reach 96 to 99 percent accuracy...

A new survey in Machine Learning provides an extensive taxonomy of attention mechanisms and more than thirty Transformer variants that...

Researchers have developed a physics-informed Kolmogorov–Arnold network that localizes multiple tunnel fire sources within about one meter and reconstructs the...

A new survey in Artificial Intelligence Review unifies graph structure learning and automated graph neural architecture search into a single...

A new hybrid 3D CNN-graph attention framework with entropy-guided fusion achieved an AUROC of 0.964 and delineated five new exploration...

A hybrid CNN–LSTM deep learning model tuned by Bayesian optimization outperformed classical and standalone neural benchmarks in quarterly GDP forecasts...

A variational autoencoder trained only on healthy motion capture data can distinguish sex, speed and gait type while generating realistic...

A new review shows how machine learning combined with spectroscopy, hyperspectral imaging and electronic sensors is enabling rapid, nondestructive prediction...