AI Glossary
The complete dictionary of Artificial Intelligence
Active Learning for Deep Learning
Data annotation optimization methodology where a deep learning model intelligently selects the most informative samples to reduce annotation effort while maximizing model performance.
Deep Bayesian Active Learning
Integration of Bayesian principles into deep neural networks to quantify epistemic uncertainty and guide sample selection for active learning.
Batch Active Learning
Variant of active learning simultaneously selecting a set of samples for annotation, optimizing computational and human costs in deep learning contexts.
Deep Ensemble Learning
Approach combining multiple deep neural network models to estimate uncertainty and improve robustness, used as a foundation for active learning strategies.
Neural Network Uncertainty
Quantification of prediction uncertainty in neural networks, distinguishing between epistemic uncertainty (lack of knowledge) and aleatoric uncertainty (inherent noise in data).
Committee-Based Sampling
Active learning strategy using a committee of deep learning models to evaluate disagreement on predictions, selecting samples with the greatest disagreement for annotation.
Deep Active Learning
Specialized domain integrating active learning techniques with deep neural network architectures to optimize annotation efficiency in massive datasets.