Glossary
There are currently 13 names in this directory beginning with the letter L.
L
Layers
Neurons are organized into layers in a neural network. The input layer receives the input data, the output layer produces the final output, and there can be one or more hidden layers in between.
Leaf node
Nodes that do not split are called leaf nodes. They represent the output or the predicted value/class for the specific subgroup of data that reaches that node.
Learning (in Bayesiant networks)
Learning in Bayesian networks involves estimating the network structure and parameters from data. This can be done using approaches such as constraint-based methods or score-based methods.
Learning rate
The learning rate is a hyperparameter that controls the step size in gradient descent. It determines how much each weak learner’s contribution affects the final ensemble. A lower learning rate makes the algorithm converge slowly but can improve generalizability, while a higher learning rate leads to faster convergence but may increase the risk of overfitting.
Lift chart
Lift is a measure of the effectiveness of a predictive model calculated as the ratio between the results obtained with and without the predictive model. The Lift chart, or Lift curve, presents the graphic representation between the lift values and response rates.
Likelihood
The likelihood in Bayesian networks represents the probability of observing specific evidence given the values of the variables in the network. It is derived from the conditional probability tables.
Line chart
A line chart is a graph that displays data points connected by straight lines, commonly used to visualize trends or changes over time or ordered categories.
Linear regression
Linear regression is a regression method model in which the target variable is interval. Linear regression is an association model.
Logistic regression
Logistic regression is a regression method in which the target variable is categorical (binary or nominal). It is a classification model.