Research Ontology
Research ontology
This website uses a small, evolving research ontology to organize publications by the areas they address, the objects they study, the tools they use, and the concrete things they produce.
Views
Research areas
- Mechanistic interpretability3 publications · graph
The study of how internal components and computations of learned models produce their behavior.
- Statistical learning theory3 publications · graph
The mathematical study of what can be learned from finite data and how performance depends on data, model classes, objectives, and algorithms.
- Training dynamics2 publications · graph
The study of how a model's parameters, representations, and computations evolve during learning.
- Adversarial robustness1 publication · graph
The study of how learning systems behave under deliberately chosen perturbations intended to cause errors or expose vulnerabilities.
- Computer vision1 publication · graph
The study of computational systems that infer structure, meaning, or action from visual data.
- Developmental interpretability1 publication · graph
The study of how interpretable representations, mechanisms, and circuits form and change during training.
- Implicit bias1 publication · graph
The study of how a learning algorithm selects among multiple solutions even without an explicit preference in the objective.
- Meta-learning1 publication · graph
The study of systems that use experience across tasks to learn how to adapt to a new task.
- Statistical modeling1 publication · graph
The construction and analysis of explicit probabilistic models for how data are generated.