Explore chapters and articles related to this topic
Machine Learning Basics
Published in Peter Wlodarczak, Machine Learning and its Applications, 2019
Predictive algorithms are often categorized into generative and discriminative models. This distinction presumes a probabilistic perspective on machine learning algorithms. Generally speaking, a generative model learns the joint probability distribution p(x, y), whereas a discriminative model learns the conditional probability distribution p(y|x), in other words, the probability of y given x. A generative algorithm models how the data was generated in order to categorize a signal. It is called generative since sampling can generate synthetic data points. Generative models ask the question: Based on my generation assumptions, which category is most likely to generate this signal? Generative models include na¨ıve Bayes, Bayesian networks, Hidden Markov models (HMM) and Markov random fields (MRF).
Understanding and Building Generative Adversarial Networks
Published in Monika Mangla, Subhash K. Shinde, Vaishali Mehta, Nonita Sharma, Sachi Nandan Mohanty, Handbook of Research on Machine Learning, 2022
In simpler terms, we can say that given a set of features X and label(s) Y, the generative approach asks the question for any given example, “Which label (Y) has the most probability to generate the features X” for example, let’s take a problem of captioning bird images, if a caption is given as “A red bird with a yellow beak sitting on a tree branch,” A suitable generative model would be capable of understanding what the content of such an image will be, while a discriminative model, on the other hand, cannot derive any meaning from the given label as it can only predict the label (Y) of a given image (X).
Convolutional and generative adversarial neural networks in manufacturing
Published in International Journal of Production Research, 2020
GANs are emerging as powerful tools for unsupervised and semi-supervised learning. A basic GAN consists of the following: A generative model (i.e. generator) generates objects. The generator does not know anything about the actual objects and learns by interacting with the discriminator. For example, a generator can produce an image.A discriminative model (i.e. discriminator) determines whether an object is actual (real, usually represented by a probability value close to 1) or fake (represented by a value close to 0).An adversarial loss (or error signal) is provided by the discriminator to the generator, thus enabling the generator to produce objects that are similar to the actual objects.
Multi-style image transfer system using conditional cycleGAN
Published in The Imaging Science Journal, 2021
Ching-Ting Tu, Hwei Jen Lin, Yihjia Tsia
Machine learning comprises mainly two types of models: discriminative model and generative model. In general, the main work of discriminative models is to learn the boundaries between categories. Common examples of discriminative model include Support Vector Machine (SVM), Logistic Regression, and Naive Bayes Classifier. On the other hand, the main work of generative models is to determine the true distribution model of the input signal, and then understand how the input signal is generated. Common examples of generative model include Hidden Markov Models, restricted Boltzmann machine, AutoEncoders (AE) [6], Variational AutoEncoder (VAE) [7], and Conditional Variational AutoEncoder (CVAE) [8].