The CAU AutoML Lab for LINC: Life, Industry, and Climate is a multidisciplinary research laboratory dedicated to advancing Automated Machine Learning or Automation by Machine Learning AutoML in the viewpoint of the theory, systems, and applications for three interconnected domains: human life, industry, and the climate. The laboratory's core mission is to design intelligent, adaptive, and scalable ML frameworks that minimize human intervention in model development while enhancing robustness, interpretability, and real-world impact. By integrating algorithmic innovation with domain-specific expertise, the lab seeks to accelerate scientific discovery, optimize complex socio-technical systems, and support evidence-based decision-making under uncertainty.

The Data & ML Group focuses on advancing fundamental methodologies in data science and ML to enable robust, adaptive, and scalable learning systems. The primary research objective of this group is to develop principled algorithms that can learn from complex, evolving, and imperfect data, with particular emphasis on data preprocessing, multi-label learning, and continual learning. Representative research activities include neural architecture search for automated model design, information-theoretic approaches based on entropy and mutual information for data representation and selection.

The Life Group focuses on applying AutoML to data derived from human life. The primary research objective of this group is to develop robust models and optimization frameworks for data observed from humans, encompassing both internal physiological signals and external behavioral patterns, with particular emphasis on interpretability, uncertainty estimation, and fairness. Representative research activities include medical image analysis to support clinical decision-making and human activity recognition for senior healthcare using wearable sensor-based data.

The Industry Group focuses on the application of AutoML across factory and business domains. The primary objective of this group is to improve productivity, reliability, and sustainability in complex industrial and commercial systems by enabling autonomous and adaptive learning from data. Core research topics include fault detection and anomaly diagnosis in manufacturing processes and products, as well as the development of recommendation systems for short-form videos, music content, and financial assets, with strong emphasis on robustness, operational safety, and explainability.

The Climate Group focuses on applying AutoML to data from climate and biological resources affected by climate change. Its research activities include developing ML models to predict, detect, and monitor Earth-scale phenomena, such as precipitation patterns, forest fires, sea ice dynamics, and other natural or anthropogenic features observable on Earth's surface. In addition, the group explores analyzing plant and animal data to develop intelligent tools that support domain experts in agriculture, environmental management, and ecological monitoring.