Wasserstein Distance
A distance metric used to measure the dissimilarity between two probability distributions.
Common AI terms beginning with W, defined for advertising professionals.
Find the Flux+Form AI glossary & dictionary to help you make sense of common AI terms. Below you can find an AI Glossary & Dictionary for “W”.
A distance metric used to measure the dissimilarity between two probability distributions.
A model that performs only slightly better than random guessing; used as a building block in boosting algorithms.
Extracting data from websites using automated tools, often to gather market intelligence.
A parameter in a neural network that determines the strength of the connection between nodes.
A regularization technique that penalizes large weights to reduce overfitting.
The strategy for setting initial values of weights before training; good initialization can speed up convergence.
Reusing the same weights across different parts of a network to reduce the number of parameters.
A neural network with many neurons in each layer, enabling it to learn diverse features.
In signal processing, a mathematical function applied to a signal to minimize edge effects before analysis.
Using AI and software to automate routine marketing tasks such as approvals, scheduling and reporting.
A dense vector representation of words capturing semantic relationships.
A popular algorithm for generating word embeddings by training a shallow neural network on large text corpora.
A feature selection technique that evaluates subsets of features by training and testing a model on them.
The dictionary defines the words. The Creative Cadence Workshop builds the practice. Eight weeks of hands-on AI training built for ad agencies and in-house creative teams.
Learn about the workshop →