← AI & Computing Sciences Division

Machine Learning Center

Advances deep learning architectures and neural network algorithms - transformers for sequential geophysical data, graph neural networks for spatial relationships, physics-informed neural networks, and federated learning - for earth science prediction and classification tasks.

Machine Learning Center Natural Language Processing Center Computer Vision Center High-Performance Computing Center Quantum Computing Center Generative AI & Foundation Models Center Reinforcement Learning & Agents Center AI Safety & Alignment Center AI for Science Center Robotics & Embodied AI Center

Research Fields 15

Statistical & Computational Learning Theory

📏 Statistical & Computational Learning Theory

Statistical learning theory studies the mathematical conditions under which a model traine…

Decision Trees & Ensemble Methods

🌳 Decision Trees & Ensemble Methods

A decision tree predicts an outcome by repeatedly splitting the data according to a sequen…

Kernel Methods & Support Vector Machines

🧿 Kernel Methods & Support Vector Machines

A kernel method solves a learning problem by working entirely in terms of similarity score…

Clustering & Unsupervised Learning

🔵 Clustering & Unsupervised Learning

Clustering partitions a collection of unlabelled data points into groups, called clusters,…

Graph Theory Foundations for Graph Neural Networks

🕸️ Graph Theory Foundations for Graph Neural Networks

A graph neural network learns representations of nodes in a graph by repeatedly having eac…

Federated & Privacy-Preserving Statistical Learning

🔐 Federated & Privacy-Preserving Statistical Learning

Federated learning trains a single shared model across many separate devices or institutio…

Self-Supervised Representation Learning & Pattern Theory

🎭 Self-Supervised Representation Learning & Pattern Theory

Self-supervised learning trains a model to build useful representations of data by solving…

Bayesian Neural Networks & Uncertainty-Aware Deep Learning

🎲 Bayesian Neural Networks & Uncertainty-Aware Deep Learning

A Bayesian neural network replaces each single fixed value in an ordinary neural network's…

Stochastic Optimization & Gradient Descent Theory

📉 Stochastic Optimization & Gradient Descent Theory

Stochastic gradient descent trains a model by repeatedly nudging its parameters in the dir…

Dimensionality Reduction & Manifold Learning

🗺️ Dimensionality Reduction & Manifold Learning

Dimensionality reduction takes data described by a large number of measured variables and …

Anomaly & Novelty Detection

🚨 Anomaly & Novelty Detection

Anomaly detection identifies data points that deviate substantially from the pattern follo…

Meta-Learning & Learning to Learn

🔁 Meta-Learning & Learning to Learn

Meta-learning trains a model not on a single fixed task but across many different but rela…

Feature Selection & Information-Theoretic Variable Selection

🎯 Feature Selection & Information-Theoretic Variable Selection

Feature selection identifies which of many available measured variables actually carry use…

Universal Function Approximation & Neural Network Architecture Theory

🧩 Universal Function Approximation & Neural Network Architecture Theory

Universal function approximation theory studies precisely which mathematical functions a n…

Active Learning & Sequential Experimental Design

🔬 Active Learning & Sequential Experimental Design

Active learning lets a model itself choose which specific unlabelled data point should be …