Tag: robustness

A new model called GA-ODE combines graph attention networks with ordinary differential equation constraints to predict opinion evolution on social ...

Researchers show that temporally correlated input noise, not model capacity, drives unstable long-horizon forecasts, and introduce NoDE, a lightweight embedding ...

Researchers have developed a self-tuning zeroing neural network with a dynamic adaptive gain and integral activation function that lets multi-agent ...

A hybrid graph neural network that combines message passing with topological persistence embeddings shows improved robustness to edge removal on ...

Researchers in Shanghai have developed a hybrid NMPC–LQR controller that lets two-wheeled self-balancing wheelchairs recover from tilts of up to ...

Technion researchers show that when multiple evolutionary paths offer equal fitness, populations deterministically drift toward flatter, more robust regions of ...

A systematic review in Applied Intelligence introduces a two-axis taxonomy that maps how large language models enhance graph neural networks ...

Researchers have developed GRaFiT, an image-text classification framework that dynamically reroutes fusion at the token, branch, and view levels to ...

Researchers have developed SLAW, a training framework that adapts label smoothing to the sharpness of the loss landscape and reweights ...

Researchers have unveiled a Trustworthy AI pipeline that embeds privacy, fairness, security, robustness, safety, and explainability across the entire development ...