X

Definition: Refers to methods and techniques in the application of artificial intelligence such that the results of the solution can be understood by human experts. It aims to address how black box decisions of AI systems are made.

Definition: A framework of syntactic theory within generative linguistics that explores the syntactic structures of a language and aims to describe these structures according to general rules. It’s relevant in computational linguistics for parsing natural language.

Definition: A highly efficient and scalable implementation of gradient boosting that is used for supervised learning problems. This algorithm builds an ensemble of models and introduces regularisation parameters to prevent overfitting, making it robust and accurate.

Definition: A markup language that defines a set of rules for encoding documents in a format that is both human-readable and machine-readable. XML is used in various AI applications to facilitate data sharing and interpretation.

Definition: A type of neural network that uses binary weights and binary inputs instead of floating-point numbers, which can drastically reduce the computational resources required for running neural networks.

Definition: A classic problem in neural network literature where the exclusive OR (XOR) logic function cannot be solved using a single layer perceptron. This problem historically demonstrated the limitations of early neural networks and led to the development of multi-layer networks.

Definition: A security vulnerability typically found in web applications. XSS enables attackers to inject client-side scripts into web pages viewed by other users. While not directly related to AI, understanding security vulnerabilities is important in the development of robust AI systems.

Definition: A language used to navigate through elements and attributes in an XML document. XPath is used in various AI contexts where XML data needs to be parsed and specific information needs to be extracted efficiently.

Definition: In computer vision and graphics, the x-coordinate refers to the horizontal component of a point or position in a 2D or 3D space. AI systems that process images or spatial data must often compute or manipulate these coordinates.

Definition: A set of software-defined hardware blocks optimised for performing AI tasks. Xilinx AI Engines are used in field-programmable gate arrays (FPGAs) to accelerate AI applications, particularly in edge devices.

Definition: Refers to the growing diversity of services available over the Internet via cloud computing as opposed to being provided locally or on-premises. In AI, this could relate to AI-as-a-Service offerings where AI tools and capabilities are provided to users through the cloud.

Definition: A term derived from the African clawed frog’s scientific name, Xenopus laevis. Xenobots are synthetic, programmable organisms designed using computer algorithms, and constructed by combining different biological tissues. This concept intersects with AI through the use of algorithms to design new forms of life.

Definition: A structured approach to implement explainable AI. It involves tools and methods designed to make AI decisions more transparent, understandable, and accountable.

Definition: A type of weight initialisation in neural networks that considers the size of the previous and next layer to maintain a variance of gradients that is neither too high nor too low throughout the training process.

Definition: A Python library for material informatics that leverages machine learning to predict material properties based on their compositions and features. This is part of a broader trend towards using AI in scientific research and development.