3,135 articles
TurboQuant
TurboQuant is a data-oblivious vector quantization algorithm from Google Research, introduced in 2025, that compresses vectors to 3 to 4 bits per coordinate near the theoretical optimum.
Turing machine
A Turing machine, also called an automatic machine or a-machine, is an abstract model of computation introduced by Alan Turing in 1936 to define what it means to compute.
Turing test in popular culture
The Turing test in popular culture is the presence of Alan Turing's 1950 machine test in fiction and film, from Blade Runner's Voight-Kampff to Her and Ex Machina.
Turkish passport
The Turkish passport, also called the Turkey passport, is the travel document issued to citizens of Türkiye in six types under the 1950 Passport Law No. 5682.
Two-phase commit protocol
The two-phase commit protocol (2PC) is an atomic commitment protocol for distributed transactions in which a coordinator asks all participants to prepare and commits only if every one agrees.
Type-2 fuzzy sets
A type-2 fuzzy set is a fuzzy set whose membership grades are themselves fuzzy, representing uncertainty about the membership degree; Lotfi Zadeh introduced them in 1975.
U-Net
U-Net is a convolutional neural network for image segmentation, published in 2015 by Olaf Ronneberger, Philipp Fischer, and Thomas Brox at the University of Freiburg.
UCI Machine Learning Repository
The UCI Machine Learning Repository is a collection of machine learning datasets maintained by the University of California, Irvine. Created in 1987, it hosts 689 datasets used by researchers worldwide.
UK Biobank
UK Biobank is a large biomedical database and prospective cohort study following 500,000 UK volunteers recruited at ages 40 to 69 between 2006 and 2010 for research into disease.
UMAP
UMAP (Uniform Manifold Approximation and Projection) is a nonlinear dimension reduction method introduced in 2018, used for visualizing high-dimensional data and faster than most t-SNE implementations.
Uncanny valley
The uncanny valley, also called bukimi no tani, is a hypothesized dip in people's affinity for figures that look almost, but not exactly, human, proposed by Masahiro Mori in 1970.
Uncertainty sampling
Uncertainty sampling is an active learning strategy where a model selects the unlabeled examples it is least certain about for labeling, introduced by Lewis and Gale in 1994.
Unconstrained binary optimization
Unconstrained binary optimization (UBO) is the class of problems optimizing an objective over binary variables with no explicit constraints, its standard form being the quadratic unconstrained binary optimization (QUBO) model.
Undersampling (machine learning)
Undersampling is a data-level technique for imbalanced classification that removes majority-class examples to balance training, using controlled methods with a set target size or cleaning methods that remove noisy or borderline examples.
Underwater computer vision
Underwater computer vision is the branch of computer vision that processes underwater images and video, applying detection, classification, segmentation, and reconstruction despite water-induced color casts, haze, and low contrast.
UNESCO Recommendation on the Ethics of AI
The UNESCO Recommendation on the Ethics of AI is a non-binding instrument adopted in Paris on 23 November 2021 by 193 member states, the first global framework on AI ethics.
Unidentified decedent
An unidentified decedent, tracked as an unidentified person, is a deceased person whose identity cannot be established despite investigation using fingerprints, dental records, and DNA.
Unified Medical Language System
The Unified Medical Language System (UMLS) is a National Library of Medicine project, begun in 1986, that integrates biomedical vocabularies so computer systems can share health data.
Uniform convergence in probability
Uniform convergence in probability is a property of a class of functions whose empirical averages converge to their true expectations simultaneously over the class, central to learning theory.
Uniform manifold approximation and projection
Uniform manifold approximation and projection (UMAP) is a nonlinear dimensionality reduction method introduced in 2018 that embeds high-dimensional data into low dimensions while preserving local neighborhoods, and is a standard tool for visualizing single-cell datasets.
UniProt
UniProt, the Universal Protein Resource, is a public database of protein sequence and functional information centered on the UniProt Knowledgebase, covering proteins from all species.
Unique games conjecture
The unique games conjecture is a conjecture in computational complexity theory, proposed by Subhash Khot in 2002, that the Unique Games constraint problem is NP-hard to approximate.
Unity Catalog (Databricks)
Unity Catalog is Databricks' unified data and AI governance catalog, open-sourced under LF AI & Data in June 2024. Uses catalog.schema.object namespace, supports Delta and Iceberg via Iceberg REST, credential vending, lineage, and auditing.
Universal approximation theorem
The universal approximation theorem, or Cybenko theorem, is a result in neural network theory showing that a one-hidden-layer network of sufficient width can approximate broad function classes to any accuracy.
Universal Dependencies
Universal Dependencies is a framework for cross-linguistically consistent morphosyntactic annotation of human language, first released in 2015 and now covering treebanks for over 190 languages.
Universal Guidelines for Artificial Intelligence
The Universal Guidelines for Artificial Intelligence are twelve principles for AI governance published in Brussels in October 2018 by The Public Voice coalition, established by EPIC.
Universal Product Code
The Universal Product Code (UPC), also called the UPC bar code, is the retail barcode symbology adopted in 1973 that lets scanners look up price and product information.
Unravel Data
Unravel Data, also called Unravel, is a Mountain View, California company (founded 2013 by Kunal Agarwal and Shivnath Babu) selling data observability and autonomous optimization, via its Arvix AI engine, for Databricks, Snowflake, BigQuery, and Cloudera.
Unscented Kalman filter
The unscented Kalman filter (UKF) is a derivative-free state-estimation algorithm for nonlinear dynamic systems that propagates a small set of sample points to approximate the state's mean and covariance.
Unsupervised change detection
Unsupervised change detection is a family of image-processing and machine-learning methods that locate differences between two co-registered images taken at different times, without labeled training samples.