Algorithme du gradient stochastiqueL'algorithme du gradient stochastique est une méthode de descente de gradient (itérative) utilisée pour la minimisation d'une fonction objectif qui est écrite comme une somme de fonctions différentiables. À la fois l'estimation statistique et l'apprentissage automatique s'intéressent au problème de la minimisation d'une fonction objectif qui a la forme d'une somme : où le paramètre qui minimise doit être estimé. Chacune des fonctions est généralement associée avec la -ème observation de l'ensemble des données (utilisées pour l'apprentissage).
Weak supervisionWeak supervision, also called semi-supervised learning, is a paradigm in machine learning, the relevance and notability of which increased with the advent of large language models due to large amount of data required to train them. It is characterized by using a combination of a small amount of human-labeled data (exclusively used in more expensive and time-consuming supervised learning paradigm), followed by a large amount of unlabeled data (used exclusively in unsupervised learning paradigm).
Generalization errorFor supervised learning applications in machine learning and statistical learning theory, generalization error (also known as the out-of-sample error or the risk) is a measure of how accurately an algorithm is able to predict outcome values for previously unseen data. Because learning algorithms are evaluated on finite samples, the evaluation of a learning algorithm may be sensitive to sampling error. As a result, measurements of prediction error on the current data may not provide much information about predictive ability on new data.
Margin (machine learning)In machine learning the margin of a single data point is defined to be the distance from the data point to a decision boundary. Note that there are many distances and decision boundaries that may be appropriate for certain datasets and goals. A margin classifier is a classifier that explicitly utilizes the margin of each example while learning a classifier. There are theoretical justifications (based on the VC dimension) as to why maximizing the margin (under some suitable constraints) may be beneficial for machine learning and statistical inferences algorithms.
Scikit-learnScikit-learn est une bibliothèque libre Python destinée à l'apprentissage automatique. Elle est développée par de nombreux contributeurs notamment dans le monde académique par des instituts français d'enseignement supérieur et de recherche comme Inria. Elle propose dans son framework de nombreuses bibliothèques d’algorithmes à implémenter, clé en main. Ces bibliothèques sont à disposition notamment des data scientists. Elle comprend notamment des fonctions pour estimer des forêts aléatoires, des régressions logistiques, des algorithmes de classification, et les machines à vecteurs de support.
Coordinate descentCoordinate descent is an optimization algorithm that successively minimizes along coordinate directions to find the minimum of a function. At each iteration, the algorithm determines a coordinate or coordinate block via a coordinate selection rule, then exactly or inexactly minimizes over the corresponding coordinate hyperplane while fixing all other coordinates or coordinate blocks. A line search along the coordinate direction can be performed at the current iterate to determine the appropriate step size.
Noyau polynomialEn apprentissage automatique, le noyau polynomial est une fonction noyau couramment utilisée avec les machines à vecteurs de support (SVMs) et d'autres modèles à noyaux. Il représente la similarité des vecteurs (échantillons d'apprentissage) dans un espace de degré polynomial plus grand que celui des variables d'origine, ce qui permet un apprentissage de modèles non-linéaires. Intuitivement, le noyau polynomial ne tient pas compte uniquement des propriétés des échantillons d'entrée afin de déterminer leur similitude, mais aussi des combinaisons de ceux-ci.
Radial basis function kernelIn machine learning, the radial basis function kernel, or RBF kernel, is a popular kernel function used in various kernelized learning algorithms. In particular, it is commonly used in support vector machine classification. The RBF kernel on two samples and x', represented as feature vectors in some input space, is defined as may be recognized as the squared Euclidean distance between the two feature vectors. is a free parameter.
LIBSVMLIBSVM and LIBLINEAR are two popular open source machine learning libraries, both developed at the National Taiwan University and both written in C++ though with a C API. LIBSVM implements the sequential minimal optimization (SMO) algorithm for kernelized support vector machines (SVMs), supporting classification and regression. LIBLINEAR implements linear SVMs and logistic regression models trained using a coordinate descent algorithm.
Platt scalingIn machine learning, Platt scaling or Platt calibration is a way of transforming the outputs of a classification model into a probability distribution over classes. The method was invented by John Platt in the context of support vector machines, replacing an earlier method by Vapnik, but can be applied to other classification models. Platt scaling works by fitting a logistic regression model to a classifier's scores. Consider the problem of binary classification: for inputs x, we want to determine whether they belong to one of two classes, arbitrarily labeled +1 and −1.
