English Index

Browse every glossary term alphabetically by its English name.

A

Accountability (Wikipedia)
説明責任せつめいせきにん
Responsibility for decisions, impacts, oversight, and remedies associated with an AI system.
Used on 1 topic page: AI Ethics
Action (Wikipedia)
行動こうどう
A choice made by the agent that can affect the environment and future rewards.
Used on 1 topic page: Reinforcement Learning
Activation function (Wikipedia)
活性化関数かっせいかかんすう
A function that transforms a neuron's combined input and introduces nonlinear behavior.
Used on 1 topic page: Neural Networks
Actor–critic (Wikipedia)
アクタークリティック
A method combining an actor that updates the policy with a critic that evaluates the actor using value estimates.
Used on 1 topic page: Reinforcement Learning
Adversarial example (Wikipedia)
敵対的てきたいてきサンプル
An input deliberately modified to cause a model to make an incorrect prediction.
Used on 1 topic page: Adversarial Machine Learning
Adversarial machine learning (Wikipedia)
敵対的機械学習てきたいてききかいがくしゅう
The study of attacks on machine-learning systems and defenses against them.
Used on 1 topic page: Adversarial Machine Learning
Agent (Wikipedia)
エージェント
The decision-making entity that observes the environment, chooses actions, and receives rewards.
Used on 1 topic page: Reinforcement Learning
AI agent (Wikipedia)
AIエージェント
An AI system that observes context, chooses actions, and works toward a goal.
Used on 1 topic page: AI Agents and Tool Use
AI ethics (Wikipedia)
AI倫理りんり
The study and practice of addressing moral issues arising from AI design, deployment, and use.
Used on 1 topic page: AI Ethics
AI governance (Wikipedia)
AIガバナンス
The structures, responsibilities, and processes used to direct and oversee AI systems.
Used on 1 topic page: AI Regulation and Policy
AI in finance (Wikipedia)
金融分野きんゆうぶんやのAI
The application of AI to financial analysis, decisions, services, and risk management.
Used on 1 topic page: AI in Finance
AI in healthcare (Wikipedia)
医療分野いりょうぶんやのAI
The application of AI to clinical care, medical operations, research, and public health.
Used on 1 topic page: AI in Healthcare
AI in robotics (Wikipedia)
ロボティクス分野ぶんやのAI
The use of AI for robot perception, decision-making, learning, and control.
Used on 1 topic page: AI in Robotics
AI regulation and policy (Wikipedia)
AI規制きせい政策せいさく
Rules and public policies governing the development, deployment, and use of AI.
Used on 1 topic page: AI Regulation and Policy
AI safety and evaluation (Wikipedia)
AIの安全性あんぜんせい評価ひょうか
The practice of measuring AI capabilities, limitations, and risks and reducing the likelihood or severity of harmful outcomes.
Used on 1 topic page: AI Safety and Evaluation
Algorithm (Wikipedia)
アルゴリズム
A defined procedure used to solve a problem or perform a computation.
Used on 1 topic page: Supervised Learning
Algorithmic bias (Wikipedia)
アルゴリズムかたよ
Systematic and unfair differences in algorithmic outcomes affecting people or groups.
Used on 1 topic page: AI Ethics
Algorithmic impact assessment (Wikipedia)
アルゴリズム影響評価えいきょうひょうか
A structured assessment of an automated system's expected effects, risks, and safeguards.
Used on 1 topic page: AI Regulation and Policy
Anomaly detection (Wikipedia)
異常いじょう検知けんち
Identifying unusual patterns or outliers in data.
Used on 1 topic page: Unsupervised Learning
Artificial intelligence (Wikipedia)
人工知能じんこうちのう
The field concerned with building machines that perform tasks associated with human intelligence.
Used on 2 topic pages: Artificial Intelligence · Machine Learning
Artificial neuron (Wikipedia)
人工じんこうニューロン
A mathematical unit that combines inputs and applies an activation function to produce an output.
Used on 1 topic page: Neural Networks
Attention (Wikipedia)
注意機構ちゅういきこう
A mechanism that assigns different weights to input elements when computing a representation.
Used on 1 topic page: Transformers and Attention
Automation bias (Wikipedia)
自動化じどうかバイアス
The tendency to favor suggestions from automated systems even when contrary evidence exists.
Used on 1 topic page: Human–AI Interaction
Autonomous system (Wikipedia)
自律じりつシステム
A system that can operate and make decisions with limited direct human control.
Used on 2 topic pages: Artificial Intelligence · Autonomous Systems

B

Backpropagation (Wikipedia)
誤差逆伝播法ごさぎゃくでんぱほう
An algorithm that computes gradients by propagating output error backward through a neural network.
Used on 1 topic page: Deep Learning
Bayesian learning (Wikipedia)
ベイズ学習がくしゅう
Learning that applies Bayes' theorem to update uncertainty about models or parameters.
Used on 1 topic page: Bayesian Learning
Benchmark (Wikipedia)
ベンチマーク
A standardized task, dataset, or measurement used to compare systems.
Used on 1 topic page: AI Safety and Evaluation
Binary classification (Wikipedia)
二値分類にちぶんるい
Classification in which each example is assigned to one of two classes.
Used on 1 topic page: Classification

C

Class label (Wikipedia)
クラスラベル
The category name or identifier predicted in a classification task.
Used on 1 topic page: Classification
Classification (Wikipedia)
分類ぶんるい
The task of assigning an input to one or more discrete categories.
Used on 1 topic page: Classification
Clinical decision support (Wikipedia)
臨床意思決定支援りんしょういしけっていしえん
Software that supplies patient-specific information to assist healthcare decisions.
Used on 1 topic page: AI in Healthcare
Clustering (Wikipedia)
クラスタリング
Grouping similar data points according to their features.
Used on 1 topic page: Unsupervised Learning
Computer vision (Wikipedia)
コンピュータビジョン
The field that enables computers to extract information from images and video.
Used on 1 topic page: Computer Vision
Context window (Wikipedia)
コンテキストウィンドウ
The amount of input and generated content a model can consider in one interaction.
Used on 1 topic page: Large Language Models
Continuous value (Wikipedia)
連続値れんぞくち
A numerical value that can vary across an interval rather than belonging to a discrete class.
Used on 1 topic page: Regression
Contrastive learning (Wikipedia)
対照学習たいしょうがくしゅう
Learning representations by bringing related examples closer and separating unrelated examples.
Used on 1 topic page: Self-Supervised Learning
Credit scoring (Wikipedia)
信用しんようスコアリング
Estimating the likelihood that a borrower will repay debt.
Used on 1 topic page: AI in Finance
Cross-validation (Wikipedia)
交差検証こうさけんしょう
A resampling method that evaluates a model across multiple training and validation partitions.
Used on 1 topic page: Model Evaluation

D

Data (Wikipedia)
データ
Recorded observations, measurements, symbols, or examples used for analysis and learning.
Used on 1 topic page: Data and Datasets
Data and datasets (Wikipedia)
データとデータセット
The examples and organized collections used for analysis and machine learning.
Used on 1 topic page: Data and Datasets
Data drift (Wikipedia)
データドリフト
A change over time in the statistical properties of data received by a deployed system.
Used on 1 topic page: Model Deployment and Monitoring
Data leakage (Wikipedia)
データリーク
The unintended use of information during training that would not be available when the model is deployed.
Used on 1 topic page: Training, Validation, and Testing
Data poisoning (Wikipedia)
データポイズニング
An attack that corrupts training data to influence a model's learned behavior.
Used on 1 topic page: Adversarial Machine Learning
Data split (Wikipedia)
データ分割ぶんかつ
The process of partitioning a dataset into distinct subsets for model development and evaluation.
Used on 1 topic page: Training, Validation, and Testing
Data visualization (Wikipedia)
データ可視化かしか
Representing data graphically to reveal patterns, trends, and insights.
Used on 1 topic page: Unsupervised Learning
Dataset (Wikipedia)
データセット
An organized collection of related data used for analysis or model development.
Used on 1 topic page: Data and Datasets
Deep learning (Wikipedia)
ディープラーニング
Machine learning based on neural networks with multiple learned representation layers.
Used on 1 topic page: Deep Learning
Deep reinforcement learning (Wikipedia)
深層強化学習しんそうきょうかがくしゅう
Reinforcement learning that uses deep neural networks to represent policies, value functions, or environment models.
Used on 1 topic page: Reinforcement Learning
Dimensionality reduction (Wikipedia)
次元じげん削減さくげん
Reducing the number of features while retaining essential structure or information.
Used on 1 topic page: Unsupervised Learning
Discount factor (Wikipedia)
割引率わりびきりつ
A value between zero and one that controls how strongly future rewards affect present decisions.
Used on 1 topic page: Reinforcement Learning
Downstream task (Wikipedia)
下流かりゅうタスク
A specific application for which a pretrained model is adapted or evaluated.
Used on 1 topic page: Foundation Models

E

Embedding (Wikipedia)

A learned numerical representation that places items in a vector space.
Used on 1 topic page: Embeddings and Representation Learning
Environment (Wikipedia)
環境かんきょう
Everything outside the agent that responds to its actions and supplies states and rewards.
Used on 1 topic page: Reinforcement Learning
Episode (Wikipedia)
エピソード
One complete sequence of interaction from an initial state until a terminal state.
Used on 1 topic page: Reinforcement Learning
Evaluation metric (Wikipedia)
評価指標ひょうかしひょう
A numerical measure used to assess a model's predictions or behavior.
Used on 1 topic page: Model Evaluation
Explainable AI (Wikipedia)
説明可能せつめいかのうなAI
Methods and practices intended to make AI behavior understandable to people.
Used on 1 topic page: Explainable AI (XAI)
Exploitation (Wikipedia)
活用かつよう
Choosing actions currently believed to produce the highest return.
Used on 1 topic page: Reinforcement Learning
Exploration (Wikipedia)
探索たんさく
Trying actions whose outcomes are uncertain in order to discover potentially better behavior.
Used on 1 topic page: Reinforcement Learning

F

Fairness (Wikipedia)
公平性こうへいせい
The study and management of unjustified differences in how AI systems affect people or groups.
Used on 1 topic page: AI Safety and Evaluation
Feature (Wikipedia)
特徴量とくちょうりょう
A measurable input property supplied to a machine-learning model.
Used on 1 topic page: Data and Datasets
Feature engineering (Wikipedia)
特徴量とくちょうりょうエンジニアリング
The process of creating or transforming model inputs to improve learning and performance.
Used on 1 topic page: Feature Engineering
Feature extraction (Wikipedia)
特徴抽出とくちょうちゅうしゅつ
Transforming raw data into informative derived features.
Used on 1 topic page: Feature Engineering
Feature importance (Wikipedia)
特徴量重要度とくちょうりょうじゅうようど
A measure of how strongly an input feature influences a model's predictions.
Used on 1 topic page: Explainable AI (XAI)
Feature selection (Wikipedia)
特徴選択とくちょうせんたく
Selecting a useful subset of available input features.
Used on 1 topic page: Feature Engineering
Federated client (Wikipedia)
連合れんごうクライアント
A participating device or organization that performs local training in federated learning.
Used on 1 topic page: Federated Learning
Federated learning (Wikipedia)
連合学習れんごうがくしゅう
Distributed learning in which participants train locally and share model updates rather than raw data.
Used on 1 topic page: Federated Learning
Few-shot learning (Wikipedia)
少数しょうすうショット学習がくしゅう
Learning or adapting from only a small number of labeled examples.
Used on 1 topic page: Meta-Learning
Fine-tuning (Wikipedia)
ファインチューニング
Further training a pretrained model on task-specific data or objectives.
Used on 1 topic page: Foundation Models
Foundation model (Wikipedia)
基盤きばんモデル
A model trained broadly at scale that can be adapted to many downstream tasks.
Used on 1 topic page: Foundation Models
Fraud detection (Wikipedia)
不正検知ふせいけんち
Identifying transactions or behavior that may involve deception or unauthorized activity.
Used on 1 topic page: AI in Finance

G

Generalization (Wikipedia)
一般化いっぱんか
A model’s ability to perform well on data it did not see during training.
Used on 1 topic page: Supervised Learning
Generative model (Wikipedia)
生成せいせいモデル
A model that learns a data distribution and can produce new samples resembling its training data.
Used on 1 topic page: Generative Models
Gradient descent (Wikipedia)
勾配降下法こうばいこうかほう
An iterative optimization method that moves parameters in the direction that reduces an objective.
Used on 1 topic page: Optimization Algorithms
Graph (Wikipedia)
グラフ
A structure consisting of entities represented as vertices and relationships represented as edges.
Used on 1 topic page: Graph Neural Networks (GNNs)
Graph neural network (Wikipedia)
グラフニューラルネットワーク
A neural network designed to learn from nodes, edges, and relationships in graph-structured data.
Used on 1 topic page: Graph Neural Networks (GNNs)
Grounding (Wikipedia)
グラウンディング
Connecting a model's output to supplied evidence, data, or observable context.
Used on 1 topic page: Retrieval-Augmented Generation (RAG)

H

Hallucination (Wikipedia)
ハルシネーション
Generated content that is unsupported, false, or inconsistent with the supplied evidence.
Used on 1 topic page: AI Safety and Evaluation
Hidden layer (Wikipedia)
かくそう
A neural-network layer located between the input and output layers.
Used on 1 topic page: Deep Learning
Human in the loop (Wikipedia)
ヒューマン・イン・ザ・ループ
A design in which people provide oversight, decisions, labels, or feedback within an automated process.
Used on 1 topic page: Human–AI Interaction
Human–AI interaction (Wikipedia)
人間にんげんとAIの相互作用そうごさよう
The study and design of communication, cooperation, and control between people and AI systems.
Used on 1 topic page: Human–AI Interaction

I

Image classification (Wikipedia)
画像がぞう分類ぶんるい
Assigning a label or category to an image.
Used on 1 topic page: Supervised Learning
Image segmentation (Wikipedia)
画像がぞうセグメンテーション
The task of assigning image pixels to meaningful regions or classes.
Used on 1 topic page: Computer Vision
Inference (Wikipedia)
推論すいろん
Using a trained model to produce an output from new input.
Used on 1 topic page: Model Deployment and Monitoring
Information retrieval (Wikipedia)
情報検索じょうほうけんさく
Finding relevant documents or records in response to an information need.
Used on 1 topic page: Retrieval-Augmented Generation (RAG)
Input data (Wikipedia)
入力にゅうりょくデータ
The raw information or features supplied to a model.
Used on 1 topic page: Supervised Learning
Intelligent agent (Wikipedia)
知的ちてきエージェント
A system that perceives an environment and acts to pursue objectives.
Used on 1 topic page: Artificial Intelligence
Interpretability (Wikipedia)
解釈可能性かいしゃくかのうせい
The degree to which a person can understand how a model produces an output.
Used on 1 topic page: Explainable AI (XAI)

L

Label (Wikipedia)
ラベル
A known category or target attached to an example in labeled data.
Used on 1 topic page: Data and Datasets
Labeled dataset (Wikipedia)
ラベルきデータセット
A collection in which each input is paired with its expected output or target.
Used on 1 topic page: Supervised Learning
Language model (Wikipedia)
言語げんごモデル
A model that assigns probabilities to sequences of language elements or predicts language content.
Used on 1 topic page: Large Language Models
Large language model (Wikipedia)
大規模言語だいきぼげんごモデル
A language model with many parameters trained on large text collections for broad language capabilities.
Used on 1 topic page: Large Language Models
Latent variable (Wikipedia)
潜在変数せんざいへんすう
An unobserved variable used by a model to explain patterns in observed data.
Used on 1 topic page: Generative Models
Localization (Wikipedia)
自己位置推定じこいちすいてい
Estimating the position and orientation of a robot or autonomous system.
Used on 1 topic page: Autonomous Systems
Loss function (Wikipedia)
損失そんしつ関数かんすう
A formula that measures the difference between predicted and actual outputs.
Used on 1 topic page: Supervised Learning

M

Machine learning (Wikipedia)
機械きかい学習がくしゅう
A field of AI in which systems learn patterns or behavior from data and experience.
Used on 4 topic pages: Artificial Intelligence · Machine Learning · Reinforcement Learning · Supervised Learning
Market segmentation (Wikipedia)
市場しじょうセグメンテーション
Dividing a market into groups with shared characteristics or behavior.
Used on 1 topic page: Unsupervised Learning
Markov Decision Process (MDP) (Wikipedia)
マルコフ決定過程けっていかてい
A mathematical framework for sequential decisions defined by states, actions, transition probabilities, rewards, and a discount factor.
Used on 1 topic page: Reinforcement Learning
Medical imaging (Wikipedia)
医用画像いようがぞう
Techniques for creating and interpreting images of the body for clinical purposes.
Used on 1 topic page: AI in Healthcare
Memory (Wikipedia)
メモリ
Stored information that an agent can retain and use across steps or interactions.
Used on 1 topic page: AI Agents and Tool Use
Message passing (Wikipedia)
メッセージパッシング
A graph-learning procedure in which nodes update their representations using information from neighboring nodes.
Used on 1 topic page: Graph Neural Networks (GNNs)
Meta-learning (Wikipedia)
メタラーニング
Learning methods that improve a system's ability to learn new tasks from limited experience.
Used on 1 topic page: Meta-Learning
Modality (Wikipedia)
モダリティ
A particular form or channel of information, such as text, vision, or audio.
Used on 1 topic page: Multimodal AI
Model (Wikipedia)
モデル
A mathematical representation that processes input data to produce predictions, classifications, or decisions.
Used on 4 topic pages: Artificial Intelligence · Machine Learning · Supervised Learning · Unsupervised Learning
Model aggregation (Wikipedia)
モデル集約しゅうやく
The process of combining model updates received from multiple federated clients.
Used on 1 topic page: Federated Learning
Model deployment (Wikipedia)
モデル展開てんかい
Making a trained model available in a production system or application.
Used on 1 topic page: Model Deployment and Monitoring
Model evaluation (Wikipedia)
モデル評価ひょうか
The systematic measurement of how well a model performs for its intended purpose.
Used on 1 topic page: Model Evaluation
Model monitoring (Wikipedia)
モデル監視かんし
Tracking a deployed model's behavior, performance, inputs, and operational health.
Used on 1 topic page: Model Deployment and Monitoring
Model-based reinforcement learning (Wikipedia)
モデルベース強化学習きょうかがくしゅう
Reinforcement learning that uses a learned or supplied model of environment dynamics to plan or improve behavior.
Used on 1 topic page: Reinforcement Learning
Model-free reinforcement learning (Wikipedia)
モデルフリー強化学習きょうかがくしゅう
Reinforcement learning that learns values or policies directly from experience without learning a model for planning.
Used on 1 topic page: Reinforcement Learning
Motion planning (Wikipedia)
動作計画どうさけいかく
Computing a feasible sequence of movements from a starting configuration to a goal.
Used on 1 topic page: AI in Robotics
Multiclass classification (Wikipedia)
クラス分類ぶんるい
Classification in which an example is selected from more than two possible classes.
Used on 1 topic page: Classification
Multimodal AI (Wikipedia)
マルチモーダルAI
AI that processes or combines more than one kind of data, such as text, images, audio, or video.
Used on 1 topic page: Multimodal AI
Multimodal embedding (Wikipedia)
マルチモーダル
A representation that places information from different modalities in a shared or aligned vector space.
Used on 1 topic page: Multimodal AI

N

Named-entity recognition (Wikipedia)
固有表現抽出こゆうひょうげんちゅうしゅつ
The task of locating and categorizing names such as people, organizations, and places in text.
Used on 1 topic page: Natural Language Processing (NLP)
Natural language processing (Wikipedia)
自然言語処理しぜんげんごしょり
The field concerned with computational processing and generation of human language.
Used on 1 topic page: Natural Language Processing (NLP)
Neural network (Wikipedia)
ニューラルネットワーク
A model composed of connected processing units arranged in layers.
Used on 1 topic page: Neural Networks

O

Object detection (Wikipedia)
物体検出ぶったいけんしゅつ
The task of locating and classifying objects within an image or video.
Used on 1 topic page: Computer Vision
Objective function (Wikipedia)
目的関数もくてきかんすう
A numerical function that an optimization process seeks to minimize or maximize.
Used on 1 topic page: Optimization Algorithms
Optimization algorithm (Wikipedia)
最適化さいてきかアルゴリズム
A procedure for finding parameter values that improve or minimize an objective.
Used on 1 topic page: Optimization Algorithms
Output label (Wikipedia)
出力しゅつりょくラベル
The expected category or value associated with an input.
Used on 1 topic page: Supervised Learning

P

Planning (Wikipedia)
計画けいかく
Selecting and ordering actions intended to move from a current state toward a goal.
Used on 1 topic page: AI Agents and Tool Use
Policy (Wikipedia)
方策ほうさく
A rule or probability distribution that determines which action an agent selects in each state.
Used on 1 topic page: Reinforcement Learning
Policy gradient (Wikipedia)
方策勾配ほうさくこうばい
A family of policy-based methods that improves a policy by following the gradient of expected return.
Used on 1 topic page: Reinforcement Learning
Policy-based method (Wikipedia)
方策ほうさくベース手法しゅほう
A method that optimizes a policy directly instead of deriving it only from learned values.
Used on 1 topic page: Reinforcement Learning
Posterior probability (Wikipedia)
事後確率じごかくりつ
A probability updated after combining prior belief with observed evidence.
Used on 1 topic page: Bayesian Learning
Prediction (Wikipedia)
予測よそく
An output produced by a trained model.
Used on 1 topic page: Supervised Learning
Prediction error (Wikipedia)
予測よそく誤差ごさ
The difference between a predicted value and its target value.
Used on 1 topic page: Supervised Learning
Pretext task (Wikipedia)
代理だいりタスク
A task constructed from unlabeled data to provide a learning signal for representation learning.
Used on 1 topic page: Self-Supervised Learning
Pretrained model (Wikipedia)
事前学習済じぜんがくしゅうずみモデル
A model whose parameters were learned before adaptation to the current task.
Used on 1 topic page: Transfer Learning
Pretraining (Wikipedia)
事前学習じぜんがくしゅう
Initial training on broad data before a model is adapted for a more specific purpose.
Used on 1 topic page: Foundation Models
Prior probability (Wikipedia)
事前確率じぜんかくりつ
A probability expressing belief before incorporating the current evidence.
Used on 1 topic page: Bayesian Learning
Probabilistic model (Wikipedia)
確率かくりつモデル
A model that represents variables and outcomes using probability distributions.
Used on 1 topic page: Probabilistic Models
Probability distribution (Wikipedia)
確率分布かくりつぶんぷ
A mathematical description of the probabilities assigned to possible outcomes.
Used on 1 topic page: Probabilistic Models

Q

Q-learning (Wikipedia)
Q学習がくしゅう
A model-free, value-based algorithm that learns the expected return for taking each action in each state.
Used on 1 topic page: Reinforcement Learning

R

Random variable (Wikipedia)
確率変数かくりつへんすう
A variable whose possible numerical values are outcomes of a random process.
Used on 1 topic page: Probabilistic Models
Red teaming (Wikipedia)
レッドチーミング
Adversarial testing that searches deliberately for failures, vulnerabilities, or harmful behavior.
Used on 1 topic page: AI Safety and Evaluation
Regression (Wikipedia)
回帰かいき
A supervised-learning task that estimates a continuous numerical target.
Used on 1 topic page: Regression
Reinforcement learning (Wikipedia)
強化学習きょうかがくしゅう
A machine-learning approach in which an agent learns behavior through interaction and feedback from rewards.
Used on 1 topic page: Reinforcement Learning
Representation learning (Wikipedia)
表現学習ひょうげんがくしゅう
Learning useful features or representations directly from data.
Used on 1 topic page: Embeddings and Representation Learning
Retrieval-augmented generation (Wikipedia)
検索拡張生成けんさくかくちょうせいせい
A generation method that retrieves external information and supplies it to a generative model as context.
Used on 1 topic page: Retrieval-Augmented Generation (RAG)
Return (Wikipedia)
収益しゅうえき
The total reward accumulated from a time step onward, usually with future rewards discounted.
Used on 1 topic page: Reinforcement Learning
Reward (Wikipedia)
報酬ほうしゅう
A numerical feedback signal that indicates the immediate desirability of an outcome.
Used on 1 topic page: Reinforcement Learning
Robot perception (Wikipedia)
ロボット知覚ちかく
The process by which a robot interprets sensor data to understand itself and its surroundings.
Used on 1 topic page: AI in Robotics
Robustness (Wikipedia)
頑健性がんけんせい
The ability of a system to maintain acceptable behavior under variation, noise, or attack.
Used on 1 topic page: AI Safety and Evaluation

S

Sampling (Wikipedia)
サンプリング
The process of drawing examples from a probability distribution or generative model.
Used on 1 topic page: Generative Models
Self-attention (Wikipedia)
自己注意機構じこちゅういきこう
Attention in which elements of one sequence attend to other elements in that same sequence.
Used on 1 topic page: Transformers and Attention
Self-supervised learning (Wikipedia)
自己教師じこきょうしあり学習がくしゅう
Learning in which supervisory signals are derived from the structure of the input data itself.
Used on 2 topic pages: Foundation Models · Self-Supervised Learning
Semantic similarity (Wikipedia)
意味的類似度いみてきるいじど
A measure of how closely two items match in meaning rather than surface form.
Used on 1 topic page: Embeddings and Representation Learning
Sensor fusion (Wikipedia)
センサフュージョン
Combining measurements from multiple sensors to obtain a more reliable estimate.
Used on 1 topic page: Autonomous Systems
Spam email detection (Wikipedia)
スパムメール検出けんしゅつ
Classifying email as spam or not spam.
Used on 1 topic page: Supervised Learning
State (Wikipedia)
状態じょうたい
A representation of the information needed to describe the environment at a particular time.
Used on 1 topic page: Reinforcement Learning
State transition (Wikipedia)
状態遷移じょうたいせんい
The change from one state to another after the agent takes an action.
Used on 1 topic page: Reinforcement Learning
Supervised learning (Wikipedia)
教師きょうしあり学習がくしゅう
A machine-learning approach in which a model is trained on labeled examples.
Used on 1 topic page: Supervised Learning
Support set (Wikipedia)
サポートセット
The small labeled example set supplied for adaptation in a few-shot learning task.
Used on 1 topic page: Meta-Learning

T

Target task (Wikipedia)
対象たいしょうタスク
The task to which transferred knowledge or a pretrained model is being applied.
Used on 1 topic page: Transfer Learning
Target value (Wikipedia)
目標もくひょう
The expected output used in supervised learning and absent from unsupervised training data.
Used on 1 topic page: Unsupervised Learning
Target variable (Wikipedia)
目的変数もくてきへんすう
The outcome that a supervised model is trained to predict.
Used on 1 topic page: Regression
Terminal state (Wikipedia)
終端状態しゅうたんじょうたい
A state that ends an episode and after which no further action is taken in that episode.
Used on 1 topic page: Reinforcement Learning
Test data (Wikipedia)
テストデータ
Held-out data used to estimate a finished model's performance on unseen examples.
Used on 1 topic page: Training, Validation, and Testing
Time step (Wikipedia)
時間じかんステップ
One discrete point in an interaction sequence at which the agent observes, acts, and receives feedback.
Used on 1 topic page: Reinforcement Learning
Token (Wikipedia)
トークン
A unit of text or other input that a model processes as one element of a sequence.
Used on 2 topic pages: Large Language Models · Transformers and Attention
Tokenization (Wikipedia)
トークン
The process of dividing input into units that a language-processing system can handle.
Used on 1 topic page: Natural Language Processing (NLP)
Tool use (Wikipedia)
ツール使用しよう
The ability of an AI system to invoke external functions, software, or services to complete work.
Used on 1 topic page: AI Agents and Tool Use
Training (Wikipedia)
訓練くんれん
The process of adjusting a model’s parameters using data.
Used on 1 topic page: Supervised Learning
Training data (Wikipedia)
訓練くんれんデータ
Examples used to fit a model's parameters or learned behavior.
Used on 2 topic pages: Machine Learning · Training, Validation, and Testing
Training, validation, and test data (Wikipedia)
訓練くんれん検証けんしょう・テストデータ
Separate data subsets used to fit a model, tune choices, and estimate final performance.
Used on 1 topic page: Training, Validation, and Testing
Transfer learning (Wikipedia)
転移学習てんいがくしゅう
Reusing knowledge learned for one task or domain to improve learning on another.
Used on 1 topic page: Transfer Learning
Transformer (Wikipedia)
トランスフォーマー
A neural-network architecture that processes relationships between tokens primarily through attention mechanisms.
Used on 2 topic pages: Large Language Models · Transformers and Attention

U

Unseen data (Wikipedia)
未見みけんデータ
Data that a model did not encounter during training.
Used on 1 topic page: Supervised Learning
Unsupervised learning (Wikipedia)
教師きょうしなし学習がくしゅう
A machine-learning approach in which a model learns patterns from unlabeled data.
Used on 1 topic page: Unsupervised Learning

V

Validation data (Wikipedia)
検証けんしょうデータ
Data used during development to compare settings or select models without fitting their parameters directly.
Used on 1 topic page: Training, Validation, and Testing
Value function (Wikipedia)
価値関数かちかんすう
A function that estimates the expected return from a state, or from taking an action in a state, under a policy.
Used on 1 topic page: Reinforcement Learning
Value-based method (Wikipedia)
価値かちベース手法しゅほう
A method that learns value estimates and derives action choices from them.
Used on 1 topic page: Reinforcement Learning
Vector (Wikipedia)
ベクトル
An ordered collection of numbers used to represent position, direction, or learned features.
Used on 1 topic page: Embeddings and Representation Learning
Vector database (Wikipedia)
ベクトルデータベース
A database designed to store embeddings and retrieve nearby vectors efficiently.
Used on 1 topic page: Retrieval-Augmented Generation (RAG)
Vision-language model (Wikipedia)
視覚言語しかくげんごモデル
A model trained to represent and reason across visual and linguistic information.
Used on 1 topic page: Multimodal AI