Training Machine Learning / Deep Learning
noun
Definition 1: A procedure in which a neural network learns to establish correspondences between input samples and correct output values [ISO/IEC 2382:2015].
Example in context: “Training a neural network is an optimization problem: we want to find the edge weight and node bias parameters that produce the best possible performance on a given task.” [Sievers et al. 2024]
Definition 2: The process of determining or improving the parameters of a machine learning model on the basis of a machine learning algorithm using training data [ISO/IEC 22989:2022].
Example in context: “Our proposed subgroup model uses patients from all subgroups for training but assigns them individual weights in the Cox partial likelihood based on their subgroup propensity score.” [Madjar et al. 2021]
Synonym: model fitting
Related terms: optimization, backpropagation, parameter estimation, fine-tuning, loss function, training data