Z

Definition: A statistical measurement that describes a value’s relationship to the mean of a group of values, measured in terms of standard deviations from the mean. In AI and machine learning, z-scores are used for normalisation of data.

Definition: A statistical test used to determine whether two population means are different when the variances are known and the sample size is large. It is often used in hypothesis testing in data science.

Definition: A type of machine learning technique where the model attempts to correctly make predictions on data that it has not explicitly seen during training. This is particularly challenging and useful for tasks like image recognition where labels for every object category cannot be provided.

Definition: A type of optimisation algorithm that does not require gradient information from the objective function. This is useful in scenarios where calculating gradients is computationally expensive or when the objective function is not differentiable.

Definition: A specification for a suite of high-level communication protocols using low-power digital radios. It is often used in Internet of Things (IoT) applications, which are increasingly integrated with AI for smarter device interactions.

Definition: A less common learning technique in neural network training where the learning direction frequently changes, potentially helping the model to escape from local minima and better explore the solution space.

Definition: A computer graphics technique managing image depth coordinates in 3D graphics, commonly used in rasterisation algorithms. While not AI-specific, understanding this is important for AI applications in game development and simulations that involve visual processing.

Definition: A unit of digital information storage that equals one sextillion bytes. As AI and big data applications grow, handling and processing data at the zettabyte scale poses significant challenges.

Definition: A cyber-attack that occurs on the same day a weakness is discovered in software. In the context of AI, predictive models can be employed to detect patterns that might indicate such vulnerabilities.

Definition: A solution by Zendesk that uses AI to automate customer service interactions, enabling companies to provide customer support through conversational interfaces.

Definition: A centralised service for maintaining configuration information, naming, providing distributed synchronisation, and providing group services. It is useful in managing distributed AI applications.

Definition: A concept from educational psychology which refers to the difference between what a learner can do without help and what he or she can achieve with guidance and encouragement from a skilled partner. AI in educational technologies often tries to operate within this zone to maximise learning efficiency.

Definition: A type of spatial analysis where statistics are calculated for zones defined by a dataset. AI and machine learning techniques can enhance these analyses, particularly in geographical information systems (GIS).

Definition: In machine vision and graphics, this is the challenge of dynamically adjusting the focus of a camera or algorithm to enhance the resolution of objects at varying distances.

Definition: A data normalisation technique where all values in a feature are scaled so that they have a mean of zero and a standard deviation of one. This is crucial in many machine learning algorithms to ensure consistent performance.