Results for "linear Gaussian"

Architecture based on self-attention and feedforward layers; foundation of modern LLMs and many multimodal models.

AI focused on interpreting images/video: classification, detection, segmentation, tracking, and 3D understanding.

Optimization problems where any local minimum is global.

Gradually increasing learning rate at training start to avoid divergence.

Allows model to attend to information from different subspaces simultaneously.

Neural networks that operate on graph-structured data by propagating information along edges.

Extension of convolution to graph domains using adjacency structure.

Generates audio waveforms from spectrograms.

Sequential data indexed by time.

Models time evolution via hidden states.

Running predictions on large datasets periodically.

Decomposes a matrix into orthogonal components; used in embeddings and compression.

Vectors with zero inner product; implies independence.

Sensitivity of a function to input perturbations.

Mathematical framework for controlling dynamic systems.

Requirement to provide explanations.

Stability proven via monotonic decrease of Lyapunov function.

Predicting protein 3D structure from sequence.

Agents optimize collective outcomes.