Machine Learning

Grokking

Grokking, also called delayed generalization, is a phenomenon in deep learning where a neural network first memorizes its training data (achieving near-perfect...

Deep Learning

Ground Truth

Ground truth is verified, correct information that serves as the authoritative reference for training and evaluating machine learning models. In supervised...

Data & Datasets

Group Attribution Bias

Group attribution bias is the tendency to assume that what is true of one member of a group is true of the entire group, or that a group's collective decision...

AI Ethics

Group Sequence Policy Optimization (GSPO)

Group Sequence Policy Optimization (GSPO) is a reinforcement learning algorithm for training large language models, introduced by the Qwen team at Alibaba in...

Reinforcement Learning

Grouped-Query Attention

Grouped-query attention (GQA) is an attention mechanism for transformer language models that partitions the query heads into a small number of groups, where...

Deep LearningTransformer Models

HNSW

Hierarchical Navigable Small World (HNSW) is an approximate nearest-neighbor search algorithm that indexes high-dimensional vectors in a multi-layer proximity...

AlgorithmsInformation Retrieval

Hallucination

Hallucination in artificial intelligence is the generation of output that is fluent, confident, and plausible but factually incorrect, fabricated, or not...

AI SafetyNatural Language Processing

Hashing

See also: Machine learning terms Hashing in machine learning is the use of hash functions to map data of arbitrary size (a word, a URL, a document, a vector)...

Information Retrieval

Heuristic

A heuristic is a practical problem-solving approach that trades optimality, completeness, accuracy, or precision for speed, producing a good-enough answer when...

Artificial Intelligence

Hidden Layer

See also: neural network, input layer, output layer, activation function A hidden layer is a layer of artificial neurons in a neural network that sits between...

Deep LearningNeural Networks

Hidden Markov Model

A Hidden Markov Model (HMM) is a statistical model of sequential data in which an unobserved (hidden) sequence of discrete states follows a Markov process, and...

Model Architecture

Hierarchical Clustering

Hierarchical clustering is an unsupervised learning method that groups data into a tree of nested clusters, building the hierarchy by repeatedly merging the...

Hinge Loss

Hinge loss is the margin-based loss function defined as max(0, 1 - y f(x)), used to train support vector machines (SVMs) and other maximum-margin classifiers,...

Training & Optimization

Holdout data

See also: Machine learning terms Holdout data is a portion of a labeled dataset that is deliberately kept out of training so it can be used later to evaluate...

Homomorphic encryption for machine learning

Homomorphic encryption for machine learning is the application of fully, somewhat, or leveled homomorphic encryption (FHE, SHE, LHE) so that a server can run...

Computer Science

HuBERT

HuBERT (Hidden-Unit BERT) is a self-supervised learning model for speech representation, introduced by researchers at Meta AI (then Facebook AI Research) in...

Speech & Audio AI

Hugging Face

Hugging Face is an artificial intelligence company and open-source platform, widely described as the "GitHub of machine learning," whose Hub hosts over 2...

AI CompaniesDeveloper Tools

Hyperparameter

See also: Machine learning terms A hyperparameter is a configuration setting in a machine learning algorithm that is fixed by the practitioner before training...

Deep LearningTraining & Optimization

Hyperparameter Tuning

Hyperparameter tuning (also called hyperparameter optimization or hyperparameter search) is the process of finding the configuration parameters that make a...

Hyperplane

A hyperplane is a flat, affine subspace of dimension n-1 embedded in an n-dimensional space, defined by the linear equation , where w is a normal vector and b...

Mathematics

I-JEPA

I-JEPA (Image-based Joint-Embedding Predictive Architecture) is a self-supervised learning method for computer vision developed by Meta AI. It was introduced...

Computer VisionMeta AI

ICML

ICML, the International Conference on Machine Learning, is one of the three top-tier academic conferences in machine learning, alongside NeurIPS and ICLR, and...

AI Events

Image Recognition

Image recognition is the field of artificial intelligence and computer vision that enables machines to identify, classify, and interpret the objects, patterns,...

Computer VisionDeep Learning

Image segmentation

Image segmentation is the computer vision task of partitioning a digital image into multiple regions by assigning every pixel a label, producing a pixel-level...

Computer VisionDeep Learning

Image-to-Text Models

See also: Multimodal Models and Tasks Image-to-text models are machine learning systems that take an image as input and produce natural language text as...

Multimodal AI

ImageNet

ImageNet is a large-scale image database of more than 14 million hand-annotated photographs organized into over 21,000 categories according to the WordNet noun...

Computer VisionDeep Learning

Imbalanced Dataset

An imbalanced dataset is a dataset used in machine learning where the classification categories are not approximately equally represented, so that one class...

Data & Datasets

Imitation Learning

Imitation learning (IL), also called learning from demonstration (LfD), is a family of machine learning methods in which an agent learns to perform a task by...

Reinforcement LearningRobotics

Implicit Bias

Implicit bias is an umbrella term that, in artificial intelligence and machine learning, refers to systematic tendencies operating below the surface of...

AI Ethics

In-Context Learning

In-context learning (ICL) is the ability of a large language model to learn a new task at inference time by conditioning on a prompt that contains a few...

Deep LearningLarge Language Models

In-Group Bias

In-group bias (also called in-group favoritism or in-group preference) is the systematic tendency to favor members of one's own social group over members of...

AI Ethics

In-set condition

An in-set condition is a split condition used inside a decision tree node that tests whether the value of a single categorical feature belongs to a specified...

InclusionAI

AI research initiative / open-source AI lab Industry February 2025 Headquarters Zhengyu He (CTO of Ant Group), Richard Bian (product and growth lead), Wu...

AI CompaniesArtificial Intelligence

Incompatibility of Fairness Metrics

The incompatibility of fairness metrics (also called the impossibility theorem of fairness or fairness trade-offs) is the proven mathematical result that...

AI Ethics

Independently and Identically Distributed (i.i.d.)

See also: Machine learning terms, Probability, Statistics, Distribution shift, Out-of-distribution detection Independently and identically distributed...

MathematicsStatistics

Individual Fairness

Individual fairness is the principle in machine learning that any two individuals who are similar with respect to a task should receive similar algorithmic...

AI Ethics

Inductive bias

Inductive bias (also called learning bias) is the set of assumptions that a learning algorithm uses to predict outputs for previously unseen inputs. Without an...

Deep LearningMathematics

Inference

AI inference is the process of running a fully trained machine learning model on new, previously unseen data to produce a prediction, classification, or...

Deep LearningMLOps

Inference optimization

Inference optimization is the set of techniques that make running a trained artificial intelligence model, especially a large language model, faster, more...

Large Language Models

Inference path

The inference path is the sequence of nodes that a single example visits as it travels from the root node of a decision tree down to a leaf node during...

Influence functions (machine learning)

An influence function is a tool for estimating how a machine learning model's predictions would change if a single training example were removed or perturbed,...

Interpretability

Information Gain

Information Gain (IG) is a measure from information theory that quantifies the reduction in entropy (uncertainty) achieved by partitioning a dataset on a...

Mathematics

Information Retrieval

Information retrieval (IR) is the science and practice of finding relevant documents, passages, or data from large unstructured collections in response to a...

Information RetrievalNatural Language Processing

Input Layer

The input layer is the first layer of a neural network: it receives the raw feature vector for each data sample and passes those values forward to the next...

Deep LearningNeural Networks

Instance

In machine learning and statistics, an instance is a single data point in a dataset: the values of one row of features and, optionally, a label [18]. It is the...

Data & Datasets

Instruction Tuning

Instruction tuning is the post-pretraining training stage in which a large language model (LLM) is fine-tuned on a curated collection of (instruction,...

Deep LearningLarge Language Models

Instruction backtranslation (Humpback)

Instruction backtranslation is a self-alignment method for generating instruction tuning data, introduced by researchers at Meta AI in the paper...

Reinforcement Learning

Integrated Gradients

Integrated Gradients (IG) is a feature-attribution method for explainable AI that explains a neural network prediction by assigning each input feature an...

Deep LearningInterpretability

Interpretability

Interpretability in artificial intelligence is the degree to which a human can understand the cause of a decision a machine learning model makes, by inspecting...

AI EthicsModel Evaluation

Iris dataset

The Iris dataset, sometimes referred to as Fisher's Iris dataset or the Iris flower dataset, is a multivariate dataset introduced by the British statistician...

AI BenchmarksData & Datasets

Item matrix

See also: Machine learning terms In collaborative filtering and matrix factorization recommender systems, the item matrix (commonly written V, sometimes Q or...

Items

See also: Machine learning terms, Machine learning terms/Recommendation Systems In recommendation systems, items are the entities that the system recommends to...

Iteration

An iteration in machine learning is a single update of a model's parameters during training, performed by processing one batch of data: the model makes...

Deep Learning

JAX

JAX is a high-performance numerical computing library developed by Google that combines a NumPy-compatible interface with composable function transformations...

Deep Learning

Joint Embedding Predictive Architecture

Joint Embedding Predictive Architecture (JEPA) is a family of self-supervised, non-generative neural network architectures proposed by Yann LeCun in his June...

Meta AIModel Architecture

Julia (programming language)

Julia is a high-level, high-performance, dynamically typed programming language built for technical computing, scientific computing, machine learning, and...

Programming Languages

JumpReLU SAE

A JumpReLU sparse autoencoder (JumpReLU SAE) is a variant of the sparse autoencoder used in mechanistic interpretability whose encoder applies a learnable...

AI Safety

K-Means

See also: Machine learning terms K-means is an unsupervised machine learning clustering algorithm that partitions a dataset into k distinct, non-overlapping...

K-Median

K-median clustering (also written k-medians) is a partitioning-based clustering algorithm that divides a dataset of objects into groups by minimizing the...

KL Divergence (Kullback-Leibler Divergence)

Kullback-Leibler divergence, often abbreviated KL divergence and written , is a measure of how one probability distribution P differs from a second reference...

MathematicsStatistics