這是幾位機器學習權威專家匯總的725個機器學習術語表，非常全面了，值得收藏！
| 英文術語 | 中文翻譯 |
|---|---|
| 0-1 Loss Function | 0-1損失函數 |
| Accept-Reject Sampling Method | 接受-拒絕抽樣法/接受-拒絕採樣法 |
| Accumulated Error Backpropagation | 累積誤差反向傳播 |
| Accuracy | 精度 |
| Acquisition Function | 採集函數 |
| Action | 動作 |
| Activation Function | 激活函數 |
| Active Learning | 主動學習 |
| Adaptive Bitrate Algorithm | 自適應比特率算法 |
| Adaptive Boosting | AdaBoost |
| Adaptive Gradient Algorithm | AdaGrad |
| Adaptive Moment Estimation Algorithm | Adam算法 |
| Adaptive Resonance Theory | 自適應諧振理論 |
| Additive Model | 加性模型 |
| Affinity Matrix | 親和矩陣 |
| Agent | 智能體 |
| Algorithm | 算法 |
| Alpha-Beta Pruning | α-β修剪法 |
| Anomaly Detection | 異常檢測 |
| Approximate Inference | 近似推斷 |
| Area Under ROC Curve | AUC |
| Artificial Intelligence | 人工智慧 |
| Artificial Neural Network | 人工神經網絡 |
| Artificial Neuron | 人工神經元 |
| Attention | 注意力 |
| Attention Mechanism | 注意力機制 |
| Attribute | 屬性 |
| Attribute Space | 屬性空間 |
| Autoencoder | 自編碼器 |
| Automatic Differentiation | 自動微分 |
| Autoregressive Model | 自回歸模型 |
| Back Propagation | 反向傳播 |
| Back Propagation Algorithm | 反向傳播算法 |
| Back Propagation Through Time | 隨時間反向傳播 |
| Backward Induction | 反向歸納 |
| Backward Search | 反向搜索 |
| Bag of Words | 詞袋 |
| Bandit | 賭博機/老虎機 |
| Base Learner | 基學習器 |
| Base Learning Algorithm | 基學習算法 |
| Baseline | 基準 |
| Batch | 批量 |
| Batch Normalization | 批量規範化 |
| Bayes Decision Rule | 貝葉斯決策準則 |
| Bayes Model Averaging | 貝葉斯模型平均 |
| Bayes Optimal Classifier | 貝葉斯最優分類器 |
| Bayes' Theorem | 貝葉斯定理 |
| Bayesian Decision Theory | 貝葉斯決策理論 |
| Bayesian Inference | 貝葉斯推斷 |
| Bayesian Learning | 貝葉斯學習 |
| Bayesian Network | 貝葉斯網/貝葉斯網絡 |
| Bayesian Optimization | 貝葉斯優化 |
| Beam Search | 束搜索 |
| Benchmark | 基準 |
| Belief Network | 信念網/信念網絡 |
| Belief Propagation | 信念傳播 |
| Bellman Equation | 貝爾曼方程 |
| Bernoulli Distribution | 伯努利分布 |
| Beta Distribution | 貝塔分布 |
| Between-Class Scatter Matrix | 類間散度矩陣 |
| BFGS | BFGS |
| Bias | 偏差/偏置 |
| Bias In Affine Function | 偏置 |
| Bias In Statistics | 偏差 |
| Bias Shift | 偏置偏移 |
| Bias-Variance Decomposition | 偏差 - 方差分解 |
| Bias-Variance Dilemma | 偏差 - 方差困境 |
| Bidirectional Recurrent Neural Network | 雙向循環神經網絡 |
| Bigram | 二元語法 |
| Bilingual Evaluation Understudy | BLEU |
| Binary Classification | 二分類 |
| Binomial Distribution | 二項分布 |
| Binomial Test | 二項檢驗 |
| Boltzmann Distribution | 玻爾茲曼分布 |
| Boltzmann Machine | 玻爾茲曼機 |
| Boosting | Boosting |
| Bootstrap Aggregating | Bagging |
| Bootstrap Sampling | 自助採樣法 |
| Bootstrapping | 自助法/自舉法 |
| Break-Event Point | 平衡點 |
| Bucketing | 分桶 |
| Calculus of Variations | 變分法 |
| Cascade-Correlation | 級聯相關 |
| Catastrophic Forgetting | 災難性遺忘 |
| Categorical Distribution | 類別分布 |
| Cell | 單元 |
| Chain Rule | 鏈式法則 |
| Chebyshev Distance | 切比雪夫距離 |
| Class | 類別 |
| Class-Imbalance | 類別不平衡 |
| Classification | 分類 |
| Classification And Regression Tree | 分類與回歸樹 |
| Classifier | 分類器 |
| Clique | 團 |
| Cluster | 簇 |
| Cluster Assumption | 聚類假設 |
| Clustering | 聚類 |
| Clustering Ensemble | 聚類集成 |
| Co-Training | 協同訓練 |
| Coding Matrix | 編碼矩陣 |
| Collaborative Filtering | 協同過濾 |
| Competitive Learning | 競爭型學習 |
| Comprehensibility | 可解釋性 |
| Computation Graph | 計算圖 |
| Computational Learning Theory | 計算學習理論 |
| Conditional Entropy | 條件熵 |
| Conditional Probability | 條件概率 |
| Conditional Probability Distribution | 條件概率分布 |
| Conditional Random Field | 條件隨機場 |
| Conditional Risk | 條件風險 |
| Confidence | 置信度 |
| Confusion Matrix | 混淆矩陣 |
| Conjugate Distribution | 共軛分布 |
| Connection Weight | 連接權 |
| Connectionism | 連接主義 |
| Consistency | 一致性 |
| Constrained Optimization | 約束優化 |
| Context Variable | 上下文變量 |
| Context Vector | 上下文向量 |
| Context Window | 上下文窗口 |
| Context Word | 上下文詞 |
| Contextual Bandit | 上下文賭博機/上下文老虎機 |
| Contingency Table | 列聯表 |
| Continuous Attribute | 連續屬性 |
| Contrastive Divergence | 對比散度 |
| Convergence | 收斂 |
| Convex Optimization | 凸優化 |
| Convex Quadratic Programming | 凸二次規劃 |
| Convolution | 卷積 |
| Convolutional Kernel | 卷積核 |
| Convolutional Neural Network | 卷積神經網絡 |
| Coordinate Descent | 坐標下降 |
| Corpus | 語料庫 |
| Correlation Coefficient | 相關係數 |
| Cosine Similarity | 餘弦相似度 |
| Cost | 代價 |
| Cost Curve | 代價曲線 |
| Cost Function | 代價函數 |
| Cost Matrix | 代價矩陣 |
| Cost-Sensitive | 代價敏感 |
| Covariance | 協方差 |
| Covariance Matrix | 協方差矩陣 |
| Critical Point | 臨界點 |
| Cross Entropy | 交叉熵 |
| Cross Validation | 交叉驗證 |
| Curse of Dimensionality | 維數災難 |
| Cutting Plane Algorithm | 割平面法 |
| Data Mining | 數據挖掘 |
| Data Set | 數據集 |
| Davidon-Fletcher-Powell | DFP |
| Decision Boundary | 決策邊界 |
| Decision Function | 決策函數 |
| Decision Stump | 決策樹樁 |
| Decision Tree | 決策樹 |
| Decoder | 解碼器 |
| Decoding | 解碼 |
| Deconvolution | 反卷積 |
| Deconvolutional Network | 反卷積網絡 |
| Deduction | 演繹 |
| Deep Belief Network | 深度信念網絡 |
| Deep Boltzmann Machine | 深度玻爾茲曼機 |
| Deep Convolutional Generative Adversarial Network | 深度卷積生成對抗網絡 |
| Deep Learning | 深度學習 |
| Deep Neural Network | 深度神經網絡 |
| Deep Q-Network | 深度Q網絡 |
| Delta-Bar-Delta | Delta-Bar-Delta |
| Denoising | 去噪 |
| Denoising Autoencoder | 去噪自編碼器 |
| Denoising Score Matching | 去躁分數匹配 |
| Density Estimation | 密度估計 |
| Density-Based Clustering | 密度聚類 |
| Derivative | 導數 |
| Determinant | 行列式 |
| Diagonal Matrix | 對角矩陣 |
| Dictionary Learning | 字典學習 |
| Dimension Reduction | 降維 |
| Directed Edge | 有向邊 |
| Directed Graphical Model | 有向圖模型 |
| Directed Separation | 有向分離 |
| Dirichlet Distribution | 狄利克雷分布 |
| Discriminative Model | 判別式模型 |
| Discriminator | 判別器 |
| Discriminator Network | 判別網絡 |
| Distance Measure | 距離度量 |
| Distance Metric Learning | 距離度量學習 |
| Distributed Representation | 分布式表示 |
| Diverge | 發散 |
| Divergence | 散度 |
| Diversity | 多樣性 |
| Diversity Measure | 多樣性度量/差異性度量 |
| Domain Adaptation | 領域自適應 |
| Dominant Strategy | 主特徵值 |
| Dominant Strategy | 占優策略 |
| Down Sampling | 下採樣 |
| Dropout | 暫退法 |
| Dropout Boosting | 暫退Boosting |
| Dropout Method | 暫退法 |
| Dual Problem | 對偶問題 |
| Dummy Node | 啞結點 |
| Dynamic Bayesian Network | 動態貝葉斯網絡 |
| Dynamic Programming | 動態規劃 |
| Early Stopping | 早停 |
| Eigendecomposition | 特徵分解 |
| Eigenvalue | 特徵值 |
| Element-Wise Product | 逐元素積 |
| Embedding | 嵌入 |
| Empirical Conditional Entropy | 經驗條件熵 |
| Empirical Distribution | 經驗分布 |
| Empirical Entropy | 經驗熵 |
| Empirical Error | 經驗誤差 |
| Empirical Risk | 經驗風險 |
| Empirical Risk Minimization | 經驗風險最小化 |
| Encoder | 編碼器 |
| Encoding | 編碼 |
| End-To-End | 端到端 |
| Energy Function | 能量函數 |
| Energy-Based Model | 基於能量的模型 |
| Ensemble Learning | 集成學習 |
| Ensemble Pruning | 集成修剪 |
| Entropy | 熵 |
| Episode | 回合 |
| Epoch | 輪 |
| Error | 誤差 |
| Error Backpropagation Algorithm | 誤差反向傳播算法 |
| Error Backpropagation | 誤差反向傳播 |
| Error Correcting Output Codes | 糾錯輸出編碼 |
| Error Rate | 錯誤率 |
| Error-Ambiguity Decomposition | 誤差－分歧分解 |
| Estimator | 估計/估計量 |
| Euclidean Distance | 歐氏距離 |
| Evidence | 證據 |
| Evidence Lower Bound | 證據下界 |
| Exact Inference | 精確推斷 |
| Example | 樣例 |
| Expectation | 期望 |
| Expectation Maximization | 期望最大化 |
| Expected Loss | 期望損失 |
| Expert System | 專家系統 |
| Exploding Gradient | 梯度爆炸 |
| Exponential Loss Function | 指數損失函數 |
| Factor | 因子 |
| Factorization | 因子分解 |
| Feature | 特徵 |
| Feature Engineering | 特徵工程 |
| Feature Map | 特徵圖 |
| Feature Selection | 特徵選擇 |
| Feature Vector | 特徵向量 |
| Featured Learning | 特徵學習 |
| Feedforward | 前饋 |
| Feedforward Neural Network | 前饋神經網絡 |
| Few-Shot Learning | 少試學習 |
| Filter | 濾波器 |
| Fine-Tuning | 微調 |
| Fluctuation | 振盪 |
| Forget Gate | 遺忘門 |
| Forward Propagation | 前向傳播/正向傳播 |
| Forward Stagewise Algorithm | 前向分步算法 |
| Fractionally Strided Convolution | 微步卷積 |
| Frobenius Norm | Frobenius 範數 |
| Full Padding | 全填充 |
| Functional | 泛函 |
| Functional Neuron | 功能神經元 |
| Gated Recurrent Unit | 門控循環單元 |
| Gated RNN | 門控RNN |
| Gaussian Distribution | 高斯分布 |
| Gaussian Kernel | 高斯核 |
| Gaussian Kernel Function | 高斯核函數 |
| Gaussian Mixture Model | 高斯混合模型 |
| Gaussian Process | 高斯過程 |
| Generalization Ability | 泛化能力 |
| Generalization Error | 泛化誤差 |
| Generalization Error Bound | 泛化誤差上界 |
| Generalize | 泛化 |
| Generalized Lagrange Function | 廣義拉格朗日函數 |
| Generalized Linear Model | 廣義線性模型 |
| Generalized Rayleigh Quotient | 廣義瑞利商 |
| Generative Adversarial Network | 生成對抗網絡 |
| Generative Model | 生成式模型 |
| Generator | 生成器 |
| Generator Network | 生成器網絡 |
| Genetic Algorithm | 遺傳算法 |
| Gibbs Distribution | 吉布斯分布 |
| Gibbs Sampling | 吉布斯採樣/吉布斯抽樣 |
| Gini Index | 基尼指數 |
| Global Markov Property | 全局馬爾可夫性 |
| Global Minimum | 全局最小 |
| Gradient | 梯度 |
| Gradient Clipping | 梯度截斷 |
| Gradient Descent | 梯度下降 |
| Gradient Descent Method | 梯度下降法 |
| Gradient Exploding Problem | 梯度爆炸問題 |
| Gram Matrix | Gram 矩陣 |
| Graph Convolutional Network | 圖卷積神經網絡/圖卷積網絡 |
| Graph Neural Network | 圖神經網絡 |
| Graphical Model | 圖模型 |
| Grid Search | 網格搜索 |
| Ground Truth | 真實值 |
| Hadamard Product | Hadamard積 |
| Hamming Distance | 漢明距離 |
| Hard Margin | 硬間隔 |
| Hebbian Rule | 赫布法則 |
| Hidden Layer | 隱藏層 |
| Hidden Markov Model | 隱馬爾可夫模型 |
| Hidden Variable | 隱變量 |
| Hierarchical Clustering | 層次聚類 |
| Hilbert Space | 希爾伯特空間 |
| Hinge Loss Function | 合頁損失函數/Hinge損失函數 |
| Hold-Out | 留出法 |
| Hyperparameter | 超參數 |
| Hyperparameter Optimization | 超參數優化 |
| Hypothesis | 假設 |
| Hypothesis Space | 假設空間 |
| Hypothesis Test | 假設檢驗 |
| Identity Matrix | 單位矩陣 |
| Imitation Learning | 模仿學習 |
| Importance Sampling | 重要性採樣 |
| Improved Iterative Scaling | 改進的疊代尺度法 |
| Incremental Learning | 增量學習 |
| Independent and Identically Distributed | 獨立同分布 |
| Indicator Function | 指示函數 |
| Individual Learner | 個體學習器 |
| Induction | 歸納 |
| Inductive Bias | 歸納偏好 |
| Inductive Learning | 歸納學習 |
| Inductive Logic Programming | 歸納邏輯程序設計 |
| Inference | 推斷 |
| Information Entropy | 信息熵 |
| Information Gain | 信息增益 |
| Inner Product | 內積 |
| Instance | 示例 |
| Internal Covariate Shift | 內部協變量偏移 |
| Inverse Matrix | 逆矩陣 |
| Inverse Resolution | 逆歸結 |
| Isometric Mapping | 等度量映射 |
| Jacobian Matrix | 雅可比矩陣 |
| Jensen Inequality | Jensen不等式 |
| Joint Probability Distribution | 聯合概率分布 |
| K-Armed Bandit Problem | k-搖臂老虎機 |
| K-Fold Cross Validation | k 折交叉驗證 |
| Karush-Kuhn-Tucker Condition | KKT條件 |
| Karush–Kuhn–Tucker | Karush–Kuhn–Tucker |
| Kernel Function | 核函數 |
| Kernel Method | 核方法 |
| Kernel Trick | 核技巧 |
| Kernelized Linear Discriminant Analysis | 核線性判別分析 |
| KL Divergence | KL散度 |
| L-BFGS | L-BFGS |
| Label | 標籤 |
| Label Space | 標記空間 |
| Lagrange Duality | 拉格朗日對偶性 |
| Lagrange Multiplier | 拉格朗日乘子 |
| Language Model | 語言模型 |
| Laplace Smoothing | 拉普拉斯平滑 |
| Laplacian Correction | 拉普拉斯修正 |
| Latent Dirichlet Allocation | 潛在狄利克雷分配 |
| Latent Semantic Analysis | 潛在語義分析 |
| Latent Variable | 潛變量/隱變量 |
| Law of Large Numbers | 大數定律 |
| Layer Normalization | 層規範化 |
| Lazy Learning | 懶惰學習 |
| Leaky Relu | 泄漏修正線性單元/泄漏整流線性單元 |
| Learner | 學習器 |
| Learning | 學習 |
| Learning By Analogy | 類比學習 |
| Learning Rate | 學習率 |
| Learning Vector Quantization | 學習向量量化 |
| Least Square Method | 最小二乘法 |
| Least Squares Regression Tree | 最小二乘回歸樹 |
| Left Singular Vector | 左奇異向量 |
| Likelihood | 似然 |
| Linear Chain Conditional Random Field | 線性鏈條件隨機場 |
| Linear Classification Model | 線性分類模型 |
| Linear Classifier | 線性分類器 |
| Linear Dependence | 線性相關 |
| Linear Discriminant Analysis | 線性判別分析 |
| Linear Model | 線性模型 |
| Linear Regression | 線性回歸 |
| Link Function | 聯繫函數 |
| Local Markov Property | 局部馬爾可夫性 |
| Local Minima | 局部極小 |
| Local Minimum | 局部極小 |
| Local Representation | 局部式表示/局部式表徵 |
| Log Likelihood | 對數似然函數 |
| Log Linear Model | 對數線性模型 |
| Log-Likelihood | 對數似然 |
| Log-Linear Regression | 對數線性回歸 |
| Logistic Function | 對數機率函數 |
| Logistic Regression | 對數機率回歸 |
| Logit | 對數機率 |
| Long Short Term Memory | 長短期記憶 |
| Long Short-Term Memory Network | 長短期記憶網絡 |
| Loopy Belief Propagation | 環狀信念傳播 |
| Loss Function | 損失函數 |
| Low Rank Matrix Approximation | 低秩矩陣近似 |
| Machine Learning | 機器學習 |
| Macron-R | 宏查全率 |
| Manhattan Distance | 曼哈頓距離 |
| Manifold | 流形 |
| Manifold Assumption | 流形假設 |
| Manifold Learning | 流形學習 |
| Margin | 間隔 |
| Marginal Distribution | 邊緣分布 |
| Marginal Independence | 邊緣獨立性 |
| Marginalization | 邊緣化 |
| Markov Chain | 馬爾可夫鏈 |
| Markov Chain Monte Carlo | 馬爾可夫鏈蒙特卡羅 |
| Markov Decision Process | 馬爾可夫決策過程 |
| Markov Network | 馬爾可夫網絡 |
| Markov Process | 馬爾可夫過程 |
| Markov Random Field | 馬爾可夫隨機場 |
| Mask | 掩碼 |
| Matrix | 矩陣 |
| Matrix Inversion | 逆矩陣 |
| Max Pooling | 最大匯聚 |
| Maximal Clique | 最大團 |
| Maximum Entropy Model | 最大熵模型 |
| Maximum Likelihood Estimation | 極大似然估計 |
| Maximum Margin | 最大間隔 |
| Mean Filed | 平均場 |
| Mean Pooling | 平均匯聚 |
| Mean Squared Error | 均方誤差 |
| Mean-Field | 平均場 |
| Memory Network | 記憶網絡 |
| Message Passing | 消息傳遞 |
| Metric Learning | 度量學習 |
| Micro-R | 微查全率 |
| Minibatch | 小批量 |
| Minimal Description Length | 最小描述長度 |
| Minimax Game | 極小極大博弈 |
| Minkowski Distance | 閔可夫斯基距離 |
| Mixture of Experts | 混合專家模型 |
| Mixture-of-Gaussian | 高斯混合 |
| Model | 模型 |
| Model Selection | 模型選擇 |
| Momentum Method | 動量法 |
| Monte Carlo Method | 蒙特卡羅方法 |
| Moral Graph | 端正圖/道德圖 |
| Moralization | 道德化 |
| Multi-Class Classification | 多分類 |
| Multi-Head Attention | 多頭注意力 |
| Multi-Head Self-Attention | 多頭自注意力 |
| Multi-Kernel Learning | 多核學習 |
| Multi-Label Learning | 多標記學習 |
| Multi-Layer Feedforward Neural Networks | 多層前饋神經網絡 |
| Multi-Layer Perceptron | 多層感知機 |
| Multinomial Distribution | 多項分布 |
| Multiple Dimensional Scaling | 多維縮放 |
| Multiple Linear Regression | 多元線性回歸 |
| Multitask Learning | 多任務學習 |
| Multivariate Normal Distribution | 多元正態分布 |
| Mutual Information | 互信息 |
| N-Gram Model | N元模型 |
| Naive Bayes Classifier | 樸素貝葉斯分類器 |
| Naive Bayes | 樸素貝葉斯 |
| Nearest Neighbor Classifier | 最近鄰分類器 |
| Negative Log Likelihood | 負對數似然函數 |
| Neighbourhood Component Analysis | 近鄰成分分析 |
| Net Input | 淨輸入 |
| Neural Network | 神經網絡 |
| Neural Turing Machine | 神經圖靈機 |
| Neuron | 神經元 |
| Newton Method | 牛頓法 |
| No Free Lunch Theorem | 沒有免費午餐定理 |
| Noise-Contrastive Estimation | 噪聲對比估計 |
| Nominal Attribute | 列名屬性 |
| Non-Convex Optimization | 非凸優化 |
| Non-Metric Distance | 非度量距離 |
| Non-Negative Matrix Factorization | 非負矩陣分解 |
| Non-Ordinal Attribute | 無序屬性 |
| Norm | 範數 |
| Normal Distribution | 正態分布 |
| Normalization | 規範化 |
| Nuclear Norm | 核範數 |
| Number of Epochs | 輪數 |
| Numerical Attribute | 數值屬性 |
| Object Detection | 目標檢測 |
| Oblique Decision Tree | 斜決策樹 |
| Occam's Razor | 奧卡姆剃刀 |
| Odds | 機率 |
| Off-Policy | 異策略 |
| On-Policy | 同策略 |
| One-Dependent Estimator | 獨依賴估計 |
| One-Hot | 獨熱 |
| Online Learning | 在線學習 |
| Optimizer | 優化器 |
| Ordinal Attribute | 有序屬性 |
| Orthogonal | 正交 |
| Orthogonal Matrix | 正交矩陣 |
| Out-Of-Bag Estimate | 包外估計 |
| Outlier | 異常點 |
| Over-Parameterized | 過度參數化 |
| Overfitting | 過擬合 |
| Oversampling | 過採樣 |
| Pac-Learnable | PAC可學習 |
| Padding | 填充 |
| Pairwise Markov Property | 成對馬爾可夫性 |
| Parallel Distributed Processing | 分布式並行處理 |
| Parameter | 參數 |
| Parameter Estimation | 參數估計 |
| Parameter Space | 參數空間 |
| Parameter Tuning | 調參 |
| Parametric ReLU | 參數化修正線性單元/參數化整流線性單元 |
| Part-Of-Speech Tagging | 詞性標註 |
| Partial Derivative | 偏導數 |
| Partially Observable Markov Decision Processes | 部分可觀測馬爾可夫決策過程 |
| Partition Function | 配分函數 |
| Perceptron | 感知機 |
| Performance Measure | 性能度量 |
| Perplexity | 困惑度 |
| Pointer Network | 指針網絡 |
| Policy | 策略 |
| Policy Gradient | 策略梯度 |
| Policy Iteration | 策略疊代 |
| Polynomial Kernel Function | 多項式核函數 |
| Pooling | 匯聚 |
| Pooling Layer | 匯聚層 |
| Positive Definite Matrix | 正定矩陣 |
| Post-Pruning | 後剪枝 |
| Potential Function | 勢函數 |
| Power Method | 冪法 |
| Pre-Training | 預訓練 |
| Precision | 查准率/準確率 |
| Prepruning | 預剪枝 |
| Primal Problem | 主問題 |
| Primary Visual Cortex | 初級視覺皮層 |
| Principal Component Analysis | 主成分分析 |
| Prior | 先驗 |
| Probabilistic Context-Free Grammar | 概率上下文無關文法 |
| Probabilistic Graphical Model | 概率圖模型 |
| Probabilistic Model | 概率模型 |
| Probability Density Function | 概率密度函數 |
| Probability Distribution | 概率分布 |
| Probably Approximately Correct | 概率近似正確 |
| Proposal Distribution | 提議分布 |
| Prototype-Based Clustering | 原型聚類 |
| Proximal Gradient Descent | 近端梯度下降 |
| Pruning | 剪枝 |
| Quadratic Loss Function | 平方損失函數 |
| Quadratic Programming | 二次規劃 |
| Quasi Newton Method | 擬牛頓法 |
| Radial Basis Function | 徑向基函數 |
| Random Forest | 隨機森林 |
| Random Sampling | 隨機採樣 |
| Random Search | 隨機搜索 |
| Random Variable | 隨機變量 |
| Random Walk | 隨機遊走 |
| Recall | 查全率/召回率 |
| Receptive Field | 感受野 |
| Reconstruction Error | 重構誤差 |
| Rectified Linear Unit | 修正線性單元/整流線性單元 |
| Recurrent Neural Network | 循環神經網絡 |
| Recursive Neural Network | 遞歸神經網絡 |
| Regression | 回歸 |
| Regularization | 正則化 |
| Regularizer | 正則化項 |
| Reinforcement Learning | 強化學習 |
| Relative Entropy | 相對熵 |
| Reparameterization | 再參數化/重參數化 |
| Representation | 表示 |
| Representation Learning | 表示學習 |
| Representer Theorem | 表示定理 |
| Reproducing Kernel Hilbert Space | 再生核希爾伯特空間 |
| Rescaling | 再縮放 |
| Reset Gate | 重置門 |
| Residual Connection | 殘差連接 |
| Residual Network | 殘差網絡 |
| Restricted Boltzmann Machine | 受限玻爾茲曼機 |
| Reward | 獎勵 |
| Ridge Regression | 嶺回歸 |
| Right Singular Vector | 右奇異向量 |
| Risk | 風險 |
| Robustness | 穩健性 |
| Root Node | 根結點 |
| Rule Learning | 規則學習 |
| Saddle Point | 鞍點 |
| Sample | 樣本 |
| Sample Complexity | 樣本複雜度 |
| Sample Space | 樣本空間 |
| Scalar | 標量 |
| Selective Ensemble | 選擇性集成 |
| Self Information | 自信息 |
| Self-Attention | 自注意力 |
| Self-Organizing Map | 自組織映射網 |
| Self-Training | 自訓練 |
| Semi-Definite Programming | 半正定規劃 |
| Semi-Naive Bayes Classifiers | 半樸素貝葉斯分類器 |
| Semi-Restricted Boltzmann Machine | 半受限玻爾茲曼機 |
| Semi-Supervised Clustering | 半監督聚類 |
| Semi-Supervised Learning | 半監督學習 |
| Semi-Supervised Support Vector Machine | 半監督支持向量機 |
| Sentiment Analysis | 情感分析 |
| Separating Hyperplane | 分離超平面 |
| Sequential Covering | 序貫覆蓋 |
| Sigmoid Belief Network | Sigmoid信念網絡 |
| Sigmoid Function | Sigmoid函數 |
| Signed Distance | 帶符號距離 |
| Similarity Measure | 相似度度量 |
| Simulated Annealing | 模擬退火 |
| Simultaneous Localization And Mapping | 即時定位與地圖構建 |
| Singular Value | 奇異值 |
| Singular Value Decomposition | 奇異值分解 |
| Skip-Gram Model | 跳元模型 |
| Smoothing | 平滑 |
| Soft Margin | 軟間隔 |
| Soft Margin Maximization | 軟間隔最大化 |
| Softmax | Softmax/軟最大化 |
| Softmax Function | Softmax函數/軟最大化函數 |
| Softmax Regression | Softmax回歸/軟最大化回歸 |
| Softplus Function | Softplus函數 |
| Span | 張成子空間 |
| Sparse Coding | 稀疏編碼 |
| Sparse Representation | 稀疏表示 |
| Sparsity | 稀疏性 |
| Specialization | 特化 |
| Splitting Variable | 切分變量 |
| Squashing Function | 擠壓函數 |
| Standard Normal Distribution | 標準正態分布 |
| State | 狀態 |
| State Value Function | 狀態值函數 |
| State-Action Value Function | 狀態-動作值函數 |
| Stationary Distribution | 平穩分布 |
| Stationary Point | 駐點 |
| Statistical Learning | 統計學習 |
| Steepest Descent | 最速下降法 |
| Stochastic Gradient Descent | 隨機梯度下降 |
| Stochastic Matrix | 隨機矩陣 |
| Stochastic Process | 隨機過程 |
| Stratified Sampling | 分層採樣 |
| Stride | 步幅 |
| Structural Risk | 結構風險 |
| Structural Risk Minimization | 結構風險最小化 |
| Subsample | 子採樣 |
| Subsampling | 下採樣 |
| Subset Search | 子集搜索 |
| Subspace | 子空間 |
| Supervised Learning | 監督學習 |
| Support Vector | 支持向量 |
| Support Vector Expansion | 支持向量展式 |
| Support Vector Machine | 支持向量機 |
| Surrogat Loss | 替代損失 |
| Surrogate Function | 替代函數 |
| Surrogate Loss Function | 代理損失函數 |
| Symbolism | 符號主義 |
| Tangent Propagation | 正切傳播 |
| Teacher Forcing | 強制教學 |
| Temporal-Difference Learning | 時序差分學習 |
| Tensor | 張量 |
| Test Error | 測試誤差 |
| Test Sample | 測試樣本 |
| Test Set | 測試集 |
| Threshold | 閾值 |
| Threshold Logic Unit | 閾值邏輯單元 |
| Threshold-Moving | 閾值移動 |
| Tied Weight | 捆綁權重 |
| Tikhonov Regularization | Tikhonov正則化 |
| Time Delay Neural Network | 時延神經網絡 |
| Time Homogenous Markov Chain | 時間齊次馬爾可夫鏈 |
| Time Step | 時間步 |
| Token | 詞元 |
| Token | 詞元 |
| Tokenization | 詞元化 |
| Tokenizer | 詞元分析器 |
| Topic Model | 話題模型 |
| Topic Modeling | 話題分析 |
| Trace | 跡 |
| Training | 訓練 |
| Training Error | 訓練誤差 |
| Training Sample | 訓練樣本 |
| Training Set | 訓練集 |
| Transductive Learning | 直推學習 |
| Transductive Transfer Learning | 直推遷移學習 |
| Transfer Learning | 遷移學習 |
| Transformer | Transformer |
| Transformer Model | Transformer模型 |
| Transpose | 轉置 |
| Transposed Convolution | 轉置卷積 |
| Trial And Error | 試錯 |
| Trigram | 三元語法 |
| Turing Machine | 圖靈機 |
| Underfitting | 欠擬合 |
| Undersampling | 欠採樣 |
| Undirected Graphical Model | 無向圖模型 |
| Uniform Distribution | 均勻分布 |
| Unigram | 一元語法 |
| Unit | 單元 |
| Universal Approximation Theorem | 通用近似定理 |
| Universal Approximator | 通用近似器 |
| Universal Function Approximator | 通用函數近似器 |
| Unknown Token | 未知詞元 |
| Unsupervised Layer-Wise Training | 無監督逐層訓練 |
| Unsupervised Learning | 無監督學習 |
| Update Gate | 更新門 |
| Upsampling | 上採樣 |
| V-Structure | V型結構 |
| Validation Set | 驗證集 |
| Validity Index | 有效性指標 |
| Value Function Approximation | 值函數近似 |
| Value Iteration | 值疊代 |
| Vanishing Gradient Problem | 梯度消失問題 |
| Vapnik-Chervonenkis Dimension | VC維 |
| Variable Elimination | 變量消去 |
| Variance | 方差 |
| Variational Autoencoder | 變分自編碼器 |
| Variational Inference | 變分推斷 |
| Vector | 向量 |
| Vector Space Model | 向量空間模型 |
| Version Space | 版本空間 |
| Viterbi Algorithm | 維特比算法 |
| Vocabulary | 詞表 |
| Warp | 線程束 |
| Weak Learner | 弱學習器 |
| Weakly Supervised Learning | 弱監督學習 |
| Weight | 權重 |
| Weight Decay | 權重衰減 |
| Weight Sharing | 權共享 |
| Weighted Voting | 加權投票 |
| Whitening | 白化 |
| Winner-Take-All | 勝者通吃 |
| Within-Class Scatter Matrix | 類內散度矩陣 |
| Word Embedding | 詞嵌入 |
| Word Sense Disambiguation | 詞義消歧 |
| Word Vector | 詞向量 |
| Zero Padding | 零填充 |
| Zero-Shot Learning | 零試學習 |
| Zipf's Law | 齊普夫定律 |