Fairness Metrics
Quantitative measures used to evaluate whether an AI system produces equitable outcomes across different groups.
Common AI terms beginning with F, 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 “F”.
Quantitative measures used to evaluate whether an AI system produces equitable outcomes across different groups.
A type of error where a model incorrectly predicts a positive case as negative, such as failing to flag a spam email.
A type of error where a model incorrectly predicts a negative case as positive, such as flagging a legitimate message as spam.
An individual measurable property or characteristic of a phenomenon being observed.
The process of creating and selecting features that improve model performance.
Metrics that quantify how much each feature contributes to a model’s predictions.
A training approach where models are trained across multiple devices or servers holding local data sets, without centralising raw data
A cycle where a model’s predictions influence the data it receives, potentially reinforcing its own biases.
A neural network architecture where information moves in one direction from input to output with no cycles.
Techniques that enable models to learn new tasks from a very small number of labelled examples.
A convolutional kernel used in image processing to extract specific patterns such as edges or textures.
Taking a pre‑trained model and training it further on a specific task or data set to specialize its capabilities.
Information collected directly from your audience or customers, used to personalize marketing and avoid dependence on third‑party data.
A loss function that down‑weights well‑classified examples to focus learning on difficult ones.
Using statistical methods or machine learning to predict future outcomes such as sales, demand or engagement.
A large, pre‑trained model that can be adapted to a wide range of tasks, such as a general‑purpose language model.
A mathematical technique that decomposes signals into constituent frequencies; used in audio and image processing.
A set of libraries and tools that provide a structure for developing AI models and applications, like TensorFlow or PyTorch.
Discovering recurring patterns in data, such as common sequences of customer actions.
A highly advanced AI model that pushes the limits of current capabilities, often requiring special governance.
Analysing each stage of the marketing funnel and applying AI to improve conversion and retention rates.
Logic that allows reasoning with degrees of truth rather than strict binary decisions; used in recommendation systems.
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 →