| absolute value rectification |
绝对值整流 |
| Accumulated error backpropagation |
累积误差逆传播 |
| accuracy |
准确率 |
| acoustic |
声学 |
| Activation Function |
激活函数 |
| AdaGrad |
AdaGrad |
| Adaptive Resonance Theory/ART |
自适应谐振理论 |
| Addictive model |
加性学习 |
| adversarial |
对抗 |
| adversarial example |
对抗样本 |
| Adversarial Networks |
对抗网络 |
| adversarial training |
对抗训练 |
| Affine Layer |
仿射层 |
| Affinity matrix |
亲和矩阵 |
| Agent |
代理 / 智能体 |
| Algorithm |
算法 |
| almost everywhere |
几乎处处 |
| almost sure |
几乎必然 |
| almost sure convergence |
几乎必然收敛 |
| Alpha-beta pruning |
α-β剪枝 |
| alternative splicing dataset |
选择性剪接数据集 |
| Ancestral Sampling |
原始采样 |
| annealed importance sampling |
退火重要采样 |
| Anomaly detection |
异常检测 |
| application-specific integratedcircuit |
专用集成电路 |
| approximate Bayesian computation |
近似贝叶斯计算 |
| approximate inference |
近似推断 |
| Approximation |
近似 |
| architecture |
架构 |
| Area Under ROC Curve／AUC |
Roc 曲线下面积 |
| Artificial General Intelligence/AGI |
通用人工智能 |
| Artificial Intelligence/AI |
人工智能 |
| artificial neural network |
人工神经网络 |
| Association analysis |
关联分析 |
| asymptotically unbiased |
渐近无偏 |
| Asynchoronous Stochastic GradientDescent |
异步随机梯度下降 |
| asynchronous |
异步 |
| Attention mechanism |
注意力机制 |
| attribute |
属性 |
| Attribute conditional independence assumption |
属性条件独立性假设 |
| Attribute space |
属性空间 |
| Attribute value |
属性值 |
| Autoencoder |
自编码器 |
| automatic differentiation |
自动微分 |
| Automatic speech recognition |
自动语音识别 |
| Automatic summarization |
自动摘要 |
| auto-regressive network |
自回归网络 |
| Average gradient |
平均梯度 |
| Average-Pooling |
平均池化 |
| ayes’ rule |
贝叶斯规则B |
| back propagate |
反向传播 |
| back propagation |
反向传播 |
| back-off |
回退 |
| backprop |
反向传播 |
| Backpropagation Through Time |
通过时间的反向传播 |
| back-propagation through time |
通过时间反向传播 |
| Backpropagation/BP |
反向传播 |
| backward propagation |
反向传播 |
| bag of words |
词袋 |
| Bagging bootstrap aggregating |
Bagging bootstrap aggregating |
| bandit |
bandit |
| Base learner |
基学习器 |
| Base learning algorithm |
基学习算法 |
| baseline |
基准 |
| batch |
批量 |
| batch normalization |
批标准化 |
| Batch Normalization/BN |
批量归一化 |
| Bayes decision rule |
贝叶斯判定准则 |
| Bayes error |
贝叶斯误差 |
| Bayes Model Averaging／BMA |
贝叶斯模型平均 |
| Bayes optimal classifier |
贝叶斯最优分类器 |
| Bayesian decision theory |
贝叶斯决策论 |
| Bayesian inference |
贝叶斯推断 |
| Bayesian network |
贝叶斯网络 |
| Bayesian probability |
贝叶斯概率 |
| Bayesian statistics |
贝叶斯统计 |
| bechmark |
基准 |
| belief network |
信念网络 |
| Bernoulli distribution |
Bernoulli 分布 |
| Between-class scatter matrix |
类间散度矩阵 |
| BFGS |
BFGS |
| Bias |
偏置 / 偏差 |
| bias in affine function |
偏置 |
| bias in statistics |
偏差 |
| biased |
有偏 |
| biased importance sampling |
有偏重要采样 |
| biass |
偏差 |
| Bias-variance decomposition |
偏差-方差分解 |
| Bias-Variance Dilemma |
偏差-方差困境 |
| Bi-directional Long-Short Term Memory/Bi-LSTM |
双向长短期记忆 |
| bigram |
二元语法 |
| Binary classification |
二分类 |
| binary relation |
二元关系 |
| binary sparse coding |
二值稀疏编码 |
| Binomial test |
二项检验 |
| Bi-partition |
二分法 |
| bit |
比特 |
| block coordinate descent |
块坐标下降 |
| block Gibbs Sampling |
块吉布斯采样 |
| Boltzmann distribution |
玻尔兹曼分布 |
| Boltzmann Machine |
玻尔兹曼机 |
| Boosting |
Boosting |
| Bootstrap sampling |
自助采样法／可重复采样／有放回采样 |
| Bootstrapping |
自助法 |
| Break-Event Point／BEP |
平衡点 |
| bridge sampling |
桥式采样 |
| broadcasting |
广播 |
| Burning-in |
磨合 |
| calculus of variations |
变分法 |
| Calibration |
校准 |
| capacity |
容量 |
| cascade |
级联 |
| Cascade-Correlation |
级联相关 |
| catastrophic forgetting |
灾难遗忘 |
| Categorical attribute |
离散属性 |
| categorical distribution |
范畴分布 |
| causal factor |
因果因子 |
| causal modeling |
因果模型 |
| ccam’s razor |
奥卡姆剃刀O |
| centered difference |
中心差分 |
| central limit theorem |
中心极限定理 |
| chain rule |
链式法则 |
| chaos |
混沌 |
| chord |
弦 |
| chordal graph |
弦图 |
| Class-conditional probability |
类条件概率 |
| Classification and regression tree/CART |
分类与回归树 |
| Classifier |
分类器 |
| Class-imbalance |
类别不平衡 |
| clip gradient |
梯度截断 |
| clipping the gradient |
截断梯度 |
| clique |
团 |
| clique potential |
团势能 |
| Closed -form |
闭式 |
| closed form solution |
闭式解 |
| Cluster |
簇/类/集群 |
| Cluster analysis |
聚类分析 |
| Clustering |
聚类 |
| Clustering ensemble |
聚类集成 |
| Co-adapting |
共适应 |
| coalesced |
级联 |
| code |
编码 |
| Coding matrix |
编码矩阵 |
| collaborative filtering |
协同过滤 |
| COLT |
国际学习理论会议 |
| column |
列 |
| column space |
列空间 |
| Committee-based learning |
基于委员会的学习 |
| common cause |
共因 |
| Competitive learning |
竞争型学习 |
| complete graph |
完全图 |
| complex cell |
复杂细胞 |
| Component learner |
组件学习器 |
| Comprehensibility |
可解释性 |
| Computation Cost |
计算成本 |
| computational graph |
计算图 |
| Computational Linguistics |
计算语言学 |
| Computer vision |
计算机视觉 |
| Concept drift |
概念漂移 |
| Concept Learning System /CLS |
概念学习系统 |
| conditional computation |
条件计算 |
| Conditional entropy |
条件熵 |
| Conditional mutual information |
条件互信息 |
| conditional probability |
条件概率 |
| conditional probability distribution |
条件概率分布 |
| Conditional Probability Table／CPT |
条件概率表 |
| Conditional random field/CRF |
条件随机场 |
| Conditional risk |
条件风险 |
| conditionally independent |
条件独立的 |
| Confidence |
置信度 |
| Confusion matrix |
混淆矩阵 |
| conjugate |
共轭 |
| conjugate directions |
共轭方向 |
| conjugate gradient |
共轭梯度 |
| Connection weight |
连接权 |
| Connectionism |
连结主义 |
| Consistency |
一致性／相合性 |
| constrained optimization |
约束优化 |
| context-specific independences |
特定环境下的独立 |
| contextual bandit |
contextual bandit |
| Contingency table |
列联表 |
| continuation method |
延拓法 |
| Continuous attribute |
连续属性 |
| contractive |
收缩 |
| contractive autoencoder |
收缩自编码器 |
| contrastive divergence |
对比散度 |
| Convergence |
收敛 |
| Conversational agent |
会话智能体 |
| Convex optimization |
凸优化 |
| Convex quadratic programming |
凸二次规划 |
| Convexity |
凸性 |
| convolution |
卷积 |
| Convolutional Boltzmann Machine |
卷积玻尔兹曼机 |
| convolutional net |
卷积网络 |
| convolutional network |
卷积网络 |
| Convolutional neural network/CNN |
卷积神经网络 |
| Co-occurrence |
同现 |
| coordinate ascent |
坐标上升 |
| coordinate descent |
坐标下降 |
| coparent |
共父 |
| correlation |
相关系数 |
| Correlation coefficient |
相关系数 |
| Cosine similarity |
余弦相似度 |
| cost |
代价 |
| Cost curve |
成本曲线 |
| Cost Function |
成本函数 |
| Cost matrix |
成本矩阵 |
| Cost-sensitive |
成本敏感 |
| covariance |
协方差 |
| covariance matrix |
协方差矩阵 |
| covariance RBM |
协方差 RBM |
| coverage |
覆盖 |
| criterion |
准则 |
| critical point |
临界点 |
| critical temperatures |
临界温度 |
| Cross entropy |
交叉熵 |
| Cross validation |
交叉验证 |
| cross-correlation |
互相关函数 |
| cross-entropy |
交叉熵 |
| Crowdsourcing |
众包 |
| cumulative function |
累积函数 |
| curriculum learning |
课程学习 |
| Curse of dimensionality |
维数灾难 |
| curvature |
曲率 |
| Cut point |
截断点 |
| Cutting plane algorithm |
割平面法 |
| cybernetics |
控制论 |
| damping |
衰减 |
| data generating distribution |
数据生成分布 |
| data generating process |
数据生成过程 |
| Data mining |
数据挖掘 |
| data parallelism |
数据并行 |
| data point |
数据点 |
| Data set |
数据集 |
| dataset |
数据集 |
| dataset augmentation |
数据集增强 |
| Decision Boundary |
决策边界 |
| Decision stump |
决策树桩 |
| Decision tree |
决策树／判定树 |
| decoder |
解码器 |
| decompose |
分解 |
| Deduction |
演绎 |
| Deep Belief Network |
深度信念网络 |
| Deep Boltzmann Machine |
深度玻尔兹曼机 |
| deep circuit |
深度回路 |
| Deep Convolutional Generative Adversarial Network/DCGAN |
深度卷积生成对抗网络 |
| deep feedforward network |
深度前馈网络 |
| deep generative model |
深度生成模型 |
| Deep learning |
深度学习 |
| deep model |
深度模型 |
| deep network |
深度网络 |
| Deep neural network/DNN |
深度神经网络 |
| Deep Q-Learning |
深度 Q 学习 |
| Deep Q-Network |
深度 Q 网络 |
| degree of belief |
信任度 |
| denoise |
去噪 |
| denoising |
去噪 |
| denoising autoencoder |
去噪自编码器 |
| denoising score matching |
去噪得分匹配 |
| Density estimation |
密度估计 |
| Density-based clustering |
密度聚类 |
| dependency |
依赖 |
| depth |
深度 |
| derivative |
导数 |
| description |
描述 |
| design matrix |
设计矩阵 |
| detailed balance |
细致平衡 |
| detector stage |
探测级 |
| deterministic |
确定性 |
| diagonal matrix |
对角矩阵 |
| Differentiable neural computer |
可微分神经计算机 |
| differential entropy |
微分熵 |
| differential equation |
微分方程 |
| dimensionality reduction |
降维 |
| Dimensionality reduction algorithm |
降维算法 |
| Dirac distribution |
Dirac 分布 |
| Diracdelta function |
Dirac delta 函数 |
| directed |
有向 |
| Directed edge |
有向边 |
| directed graphical model |
有向图模型 |
| Directed Model |
有向模型 |
| directional derivative |
方向导数 |
| Disagreement measure |
不合度量 |
| Discriminative model |
判别模型 |
| Discriminator |
判别器 |
| discriminator network |
判别器网络 |
| Distance measure |
距离度量 |
| Distance metric learning |
距离度量学习 |
| distributed representation |
分布式表示 |
| Distribution |
分布 |
| Divergence |
散度 |
| Diversity measure |
多样性度量／差异性度量 |
| DNN |
深度神经网络 |
| Domain adaption |
领域自适应 |
| dot product |
点积 |
| double backprop |
双反向传播 |
| doubly block circulant matrix |
双重分块循环矩阵 |
| Downsampling |
下采样/降采样 |
| Dropout |
Dropout |
| Dropout Boosting |
Dropout Boosting |
| Dropout fast dropout |
快速 |
| d-separation |
d-分离 |
| D-separation （Directed separation） |
有向分离 |
| Dual problem |
对偶问题 |
| Dummy node |
哑结点 |
| Dynamic Fusion |
动态融合 |
| Dynamic programming |
动态规划 |
| dynamic structure |
动态结构 |
| early stopping |
提前终止 |
| echo state network |
回声状态网络 |
| effective capacity |
有效容量 |
| eibniz’s rule |
莱布尼兹法则L |
| eigendecomposition |
特征分解 |
| eigenvalue |
特征值 |
| Eigenvalue decomposition |
特征值分解 |
| eigenvector |
特征向量 |
| elementary basis vectors |
基本单位向量 |
| element-wise product |
元素对应乘积 |
| Embedding |
嵌入 |
| Emotional analysis |
情绪分析 |
| Empirical conditional entropy |
经验条件熵 |
| empirical distribution |
经验分布 |
| Empirical entropy |
经验熵 |
| Empirical error |
经验误差 |
| empirical frequency |
经验频率 |
| Empirical risk |
经验风险 |
| empirical risk minimization |
经验风险最小化 |
| encoder |
编码器 |
| End-to-End |
端到端 |
| energy function |
能量函数 |
| Energy-based model |
基于能量的模型 |
| ensemble |
集成 |
| Ensemble learning |
集成学习 |
| Ensemble pruning |
集成修剪 |
| epoch |
轮 |
| epochs |
轮数 |
| equality constraint |
等式约束 |
| Equilibrium Distribution |
均衡分布 |
| equivariance |
等变 |
| equivariant representations |
等变表示 |
| error bar |
误差条 |
| Error Correcting Output Codes／ECOC |
纠错输出码 |
| error function |
误差函数 |
| error metric |
误差度量 |
| Error rate |
错误率 |
| Error-ambiguity decomposition |
误差-分歧分解 |
| estimator |
估计量 |
| Euclidean distance |
欧氏距离 |
| Euclidean norm |
欧几里得范数 |
| Euler-LagrangeEquation |
欧拉-拉格朗日方程 |
| evidence lower bound |
证据下界 |
| Evolutionary computation |
演化计算 |
| ewton’s method |
牛顿法N |
| example |
样本 |
| excess error |
额外误差 |
| expectation |
期望 |
| expectation maximization |
期望最大化 |
| expectation step |
E步 |
| Expectation-Maximization |
期望最大化 |
| Expected loss |
期望损失 |
| expected value |
期望值 |
| experience E |
经验 |
| expert network |
专家网络 |
| explaining away |
相消解释 |
| explaining away effect |
相消解释作用 |
| explanatory factort |
解释因子 |
| exploding gradient |
梯度爆炸 |
| Exploding Gradient Problem |
梯度爆炸问题 |
| exploitation |
利用 |
| exploration |
探索 |
| exponential distribution |
指数分布 |
| Exponential loss function |
指数损失函数 |
| Extreme Learning Machine/ELM |
超限学习机 |
| factor |
因子 |
| factor analysis |
因子分析 |
| factor graph |
因子图 |
| factorial |
因子 |
| Factorization |
因子分解/分解 |
| factorized |
分解的 |
| factors of variation |
变差因素 |
| False negative |
假负类 |
| False positive |
假正类 |
| False Positive Rate/FPR |
假正例率 |
| fast persistent contrastivedivergence |
快速持续性对比散度 |
| feasible |
可行 |
| feature |
特征 |
| Feature engineering |
特征工程 |
| feature extractor |
特征提取器 |
| feature map |
特征映射 |
| feature selection |
特征选择 |
| Feature vector |
特征向量 |
| Featured Learning |
特征学习 |
| feedback |
反馈 |
| feedforward |
前向 |
| feedforward classifier |
前馈分类器 |
| feedforward network |
前馈网络 |
| Feedforward Neural Networks/FNN |
前馈神经网络 |
| field programmable gated array |
现场可编程门阵列 |
| fine-tune |
精调 |
| Fine-tuning |
微调 |
| finite difference |
有限差分 |
| first layer |
第一层 |
| fixed point equation |
不动点方程 |
| fixed-point arithmetic |
定点运算 |
| flip |
翻转 |
| Flipping output |
翻转法 |
| float-point arithmetic |
浮点运算 |
| Fluctuation |
震荡 |
| forget gate |
遗忘门 |
| forward mode accumulation |
前向模式累加 |
| forward propagation |
前向传播 |
| Forward stagewise algorithm |
前向分步算法 |
| Fourier transform |
傅立叶变换 |
| fovea |
中央凹 |
| free energy |
自由能 |
| Frequentist |
频率主义学派 |
| frequentist probability |
频率派概率 |
| frequentist statistics |
频率派统计 |
| Frobenius norm |
Frobenius 范数 |
| F-score |
F 分数 |
| full |
全 |
| Full-rank matrix |
满秩矩阵 |
| functional |
泛函 |
| functional derivative |
泛函导数 |
| Functional neuron |
功能神经元 |
| Gaborfunction |
Gabor 函数 |
| Gain ratio |
增益率 |
| Game theory |
博弈论 |
| Gammadistribution |
Gamma 分布 |
| gated |
门控 |
| gated recurrent net |
门控循环网络 |
| gated recurrent unit |
门控循环单元 |
| gater |
选通器 |
| Gaussian distribution |
高斯分布 |
| Gaussian kernel |
高斯核 |
| Gaussian kernel function |
高斯核函数 |
| Gaussian Mixture Model |
高斯混合模型 |
| Gaussian mixtures |
高斯混合体 |
| Gaussian output distribution |
高斯输出分布 |
| Gaussian-BernoulliRBM |
Gaussian-Bernoulli RBM |
| General Problem Solving |
通用问题求解 |
| Generalization |
泛化 |
| Generalization error |
泛化误差 |
| Generalization error bound |
泛化误差上界 |
| generalize |
泛化 |
| generalized function |
广义函数 |
| Generalized Lagrange function |
广义拉格朗日函数 |
| Generalized linear model |
广义线性模型 |
| generalized pseudolikelihood |
广义伪似然 |
| generalized pseudolikelihoodestimator |
广义伪似然估计 |
| Generalized Rayleigh quotient |
广义瑞利商 |
| generalized score matching |
广义得分匹配 |
| generalizedLagrange function |
广义Lagrange 函数 |
| generative adversarial framework |
生成式对抗框架 |
| generative adversarial network |
生成式对抗网络 |
| Generative Adversarial Networks/GAN |
生成对抗网络 |
| Generative Model |
生成模型 |
| generative modeling |
生成式建模 |
| generative moment matchingnetwork |
生成矩匹配网络 |
| generative stochastic network |
生成随机网络 |
| Generator |
生成器 |
| generator network |
生成器网络 |
| Genetic Algorithm/GA |
遗传算法 |
| Gibbs distribution |
吉布斯分布 |
| Gibbs sampling |
吉布斯采样 |
| Gibbs steps |
吉布斯步数 |
| GibbsSampling |
Gibbs 采样 |
| Gini index |
基尼指数 |
| Global contrast normalization |
全局对比度归一化 |
| global minima |
全局极小值 |
| Global minimum |
全局最小 |
| Global Optimization |
全局优化 |
| GPU general purpose GPU |
通用 |
| gradient |
梯度 |
| gradient ascent |
梯度上升 |
| Gradient boosting |
梯度提升 |
| gradient clipping |
梯度截断 |
| Gradient Descent |
梯度下降 |
| Graph theory |
图论 |
| graphical model |
图模型 |
| Graphics Processing Unit |
图形处理器 |
| greedy |
贪心 |
| greedy algorithm |
贪心算法 |
| greedy layer-wise pretraining |
贪心逐层预训练 |
| greedy layer-wise training |
贪心逐层训练 |
| greedy layer-wise unsupervisedpretraining |
贪心逐层无监督预训练 |
| greedy supervised pretraining |
贪心监督预训练 |
| greedy unsupervisedpretraining |
贪心无监督预训练 |
| grid search |
网格搜索 |
| Ground-truth |
真相／真实 |
| Hadamardproduct |
Hadamard 乘积 |
| Hamming distance |
汉明距离 |
| Hard margin |
硬间隔 |
| hard mixture of experts |
硬专家混合体 |
| hard tanh |
硬双曲正切函数 |
| Hard voting |
硬投票 |
| Harmonic mean |
调和平均 |
| harmonium |
簧风琴 |
| Harris Chain |
哈里斯链 |
| Helmholtzmachine |
Helmholtz 机 |
| Hesse matrix |
海塞矩阵 |
| Hessian |
Hessian |
| heteroscedastic |
异方差 |
| Hidden dynamic model |
隐动态模型 |
| hidden layer |
隐藏层 |
| Hidden Markov Model/HMM |
隐马尔可夫模型 |
| hidden unit |
隐藏单元 |
| hidden variable |
隐藏变量 |
| Hierarchical clustering |
层次聚类 |
| Hilbert space |
希尔伯特空间 |
| hill climbing |
爬山 |
| Hinge loss function |
合页损失函数 |
| Hold-out |
留出法 |
| Homogeneous |
同质 |
| Hybrid computing |
混合计算 |
| hyperparameter |
超参数 |
| hyperparameter optimization |
超参数优化 |
| Hypothesis |
假设 |
| hypothesis space |
假设空间 |
| Hypothesis test |
假设验证 |
| i.i.d. assumption |
独立同分布假设 |
| ICA topographic ICA |
地质 |
| ICML |
国际机器学习会议 |
| identically distributed |
同分布的 |
| identifiable |
可辨认的 |
| identity matrix |
单位矩阵 |
| ill conditioning |
病态 |
| immorality |
不道德 |
| Importance Sampling |
重要采样 |
| Improved iterative scaling/IIS |
改进的迭代尺度法 |
| Incremental learning |
增量学习 |
| independent |
相互独立的 |
| Independent and identically distributed/i.i.d. |
独立同分布 |
| Independent Component Analysis/ICA |
独立成分分析 |
| independent identicallydistributed |
独立同分布 |
| independent subspace analysis |
独立子空间分析 |
| index of matrix |
索引 |
| Indicator function |
指示函数 |
| Individual learner |
个体学习器 |
| Induction |
归纳 |
| Inductive bias |
归纳偏好 |
| Inductive learning |
归纳学习 |
| Inductive Logic Programming／ILP |
归纳逻辑程序设计 |
| inequality constraint |
不等式约束 |
| inference |
推断 |
| infinite |
无限 |
| Information entropy |
信息熵 |
| Information gain |
信息增益 |
| information retrieval |
信息检索 |
| inner product |
内积 |
| input |
输入 |
| input distribution |
输入分布 |
| Input layer |
输入层 |
| Insensitive loss |
不敏感损失 |
| Inter-cluster similarity |
簇间相似度 |
| International Conference for Machine Learning/ICML |
国际机器学习大会 |
| intervention query |
干预查询 |
| Intra-cluster similarity |
簇内相似度 |
| Intrinsic value |
固有值 |
| invariant |
不变 |
| invert |
求逆 |
| Isomap |
Isomap |
| Isometric Mapping/Isomap |
等度量映射 |
| Isotonic regression |
等分回归 |
| isotropic |
各向同性 |
| Iterative Dichotomiser |
迭代二分器 |
| Jacobian |
Jacobian |
| Jacobianmatrix |
Jacobian 矩阵 |
| joint probability distribution |
联合概率分布 |
| KarushKuhnTucker |
KarushKuhnTucker |
| kernel function |
核函数 |
| kernel machine |
核机器 |
| Kernel method |
核方法 |
| Kernel trick |
核技巧 |
| Kernelized Linear Discriminant Analysis／KLDA |
核线性判别分析 |
| K-fold cross validation |
k 折交叉验证／k 倍交叉验证 |
| KLdivergence |
KL散度 |
| K-Means Clustering |
K – 均值聚类 |
| K-Nearest Neighbours Algorithm/KNN |
K近邻算法 |
| Knowledge base |
知识库 |
| knowledge graph |
知识图谱 |
| Knowledge Representation |
知识表征 |
| Krylovmethod |
Krylov 方法 |
| Kullback-Leibler(KL) divergence |
KL 散度 |
| label |
标签 |
| Label space |
标记空间 |
| labeled |
标注 |
| Lagrange duality |
拉格朗日对偶性 |
| Lagrange multiplier |
拉格朗日乘子 |
| Lagrangian generalized Lagrangian |
广义 |
| language model |
语言模型 |
| Laplace smoothing |
拉普拉斯平滑 |
| Laplacedistribution |
Laplace 分布 |
| Laplacian correction |
拉普拉斯修正 |
| large learning step |
大学习步骤 |
| latent |
潜在 |
| Latent Dirichlet Allocation |
隐狄利克雷分布 |
| latent layer |
潜层 |
| Latent semantic analysis |
潜在语义分析 |
| Latent variable |
隐/潜变量 |
| Law of large number |
大数定理 |
| layer-wise |
逐层的 |
| Lazy learning |
懒惰学习 |
| L-BFGS |
L-BFGS |
| Leaky ReLU |
渗漏整流线性单元 |
| leaky unit |
渗漏单元 |
| learned |
学成 |
| learned approximate inference |
学习近似推断 |
| Learner |
学习器 |
| Learning by analogy |
类比学习 |
| Learning rate |
学习率 |
| Learning Vector Quantization/LVQ |
学习向量量化 |
| Least squares regression tree |
最小二乘回归树 |
| Leave-One-Out/LOO |
留一法 |
| Lebesgue-integrable |
勒贝格可积 |
| left eigenvector |
左特征向量 |
| left singular vector |
左奇异向量 |
| likelihood |
似然 |
| line search |
线搜索 |
| linear auto-regressive network |
线性自回归网络 |
| linear chain conditional random field |
线性链条件随机场 |
| linear classifier |
线性分类器 |
| linear combination |
线性组合 |
| linear dependence |
线性相关 |
| Linear Discriminant Analysis／LDA |
线性判别分析 |
| linear factor model |
线性因子模型 |
| Linear model |
线性模型 |
| Linear Regression |
线性回归 |
| linear threshold units |
线性阀值单元 |
| linearly independent |
线性无关 |
| Link function |
联系函数 |
| link prediction |
链接预测 |
| linked importance sampling |
链接重要采样 |
| Lipschitz |
Lipschitz |
| Lipschitzconstant |
Lipschitz 常数 |
| Lipschitzcontinuous |
Lipschitz 连续 |
| liquid state machine |
流体状态机 |
| local conditional probabilitydistribution |
局部条件概率分布 |
| local constancy prior |
局部不变性先验 |
| local contrast normalization |
局部对比度归一化 |
| local descent |
局部下降 |
| local kernel |
局部核 |
| Local Markov property |
局部马尔可夫性 |
| local maxima |
局部极大值 |
| local maximum |
局部极大点 |
| local minima |
局部极小值 |
| Local minimum |
局部最小 |
| Log likelihood |
对数似然 |
| Log odds／logit |
对数几率 |
| logarithmic scale |
对数尺度 |
| Logistic Regression |
Logistic 回归 |
| logistic sigmoid |
logistic sigmoid |
| logit |
分对数 |
| Log-likelihood |
对数似然 |
| log-linear model |
对数线性模型 |
| Log-linear regression |
对数线性回归 |
| Long-Short Term Memory/LSTM |
长短期记忆 |
| long-term dependency |
长期依赖 |
| loop |
环 |
| loopy belief propagation |
环状信念传播 |
| loss |
损失 |
| Loss function |
损失函数 |
| machine learning |
机器学习 |
| machine learning model |
机器学习模型 |
| Machine translation/MT |
机器翻译 |
| Macron-P |
宏查准率 |
| Macron-R |
宏查全率 |
| main diagonal |
主对角线 |
| Majority voting |
绝对多数投票法 |
| manifold |
流形 |
| Manifold assumption |
流形假设 |
| manifold hypothesis |
流形假设 |
| Manifold learning |
流形学习 |
| manifold tangent classifier |
流形正切分类器 |
| Margin theory |
间隔理论 |
| Marginal distribution |
边际分布 |
| Marginal independence |
边际独立性 |
| marginal probability distribution |
边缘概率分布 |
| Marginalization |
边际化 |
| Markov Chain Monte Carlo |
马尔可夫链蒙特卡罗 |
| Markov Chain Monte Carlo/MCMC |
马尔可夫链蒙特卡罗方法 |
| Markov Chain xv |
马尔可夫链 |
| Markov network |
马尔可夫网络 |
| Markov Random Field |
马尔可夫随机场 |
| mask |
掩码 |
| matrix |
矩阵 |
| matrix inversion |
矩阵逆 |
| matrix product |
矩阵乘积 |
| max norm |
最大范数 |
| max pooling |
最大池化 |
| maxima |
极大值 |
| Maximal clique |
最大团 |
| maximizationstep |
M 步 |
| Maximum A Posteriori |
最大后验 |
| maximum likelihood |
最大似然 |
| maximum likelihood estimation |
最大似然估计 |
| Maximum Likelihood Estimation/MLE |
极大似然估计／极大似然法 |
| Maximum margin |
最大间隔 |
| maximum mean discrepancy |
最大平均偏差 |
| Maximum weighted spanning tree |
最大带权生成树 |
| maxout |
maxout |
| maxoutunit |
maxout 单元 |
| Max-Pooling |
最大池化 |
| mean absolute error |
平均绝对误差 |
| mean and covariance RBM |
均值和协方差RBM |
| Mean squared error |
均方误差 |
| mean-covariance restricted Boltzmann machine |
均值-协方差RBM |
| meanfield |
均匀场 |
| mean-field |
均值场 |
| meanproduct of Student t-distribution |
学生t 分布均值乘积 |
| measure theory |
测度论 |
| measure zero |
零测度 |
| memory network |
记忆网络 |
| message passing |
信息传输 |
| Meta-learner |
元学习器 |
| Metric learning |
度量学习 |
| Micro-P |
微查准率 |
| Micro-R |
微查全率 |
| minibatch |
小批量 |
| minibatch stochastic |
小批量随机 |
| minima |
极小值 |
| Minimal Description Length/MDL |
最小描述长度 |
| Minimax game |
极小极大博弈 |
| minimum |
极小点 |
| Misclassification cost |
误分类成本 |
| Mixing |
混合 |
| Mixing Time |
混合时间 |
| mixture density network |
混合密度网络 |
| mixture distribution |
混合分布 |
| Mixture of experts |
混合专家/专家混合体 |
| modality |
模态 |
| mode |
峰值 |
| model |
模型 |
| model averaging |
模型平均 |
| model compression |
模型压缩 |
| model identifiability |
模型可辨识性 |
| model parallelism |
模型并行 |
| moment |
矩 |
| moment matching |
矩匹配 |
| Momentum |
动量 |
| Monte Carlo |
蒙特卡罗 |
| Moore-Penrosepseudoinverse |
Moore-Penrose 伪逆 |
| Moral graph |
道德图／端正图 |
| moralization |
道德化 |
| moralized graph |
道德图 |
| Multi-class classification |
多分类 |
| Multi-document summarization |
多文档摘要 |
| Multi-layer feedforward neural networks |
多层前馈神经网络 |
| multilayer perceptron |
多层感知机 |
| Multilayer Perceptron/MLP |
多层感知器 |
| multimodal |
多峰值 |
| Multimodal learning |
多模态学习 |
| multinomial distribution |
多项式分布 |
| multinoullidistribution |
Multinoulli 分布 |
| Multiple Dimensional Scaling |
多维缩放 |
| Multiple linear regression |
多元线性回归 |
| multi-prediction deep Boltzmannmachine |
多预测深度玻尔兹曼机 |
| Multi-response Linear Regression ／MLR |
多响应线性回归 |
| multitask learning |
多任务学习 |
| multivariate normal distribution |
多维正态分布 |
| Mutual information |
互信息 |
| Naive bayes |
朴素贝叶斯 |
| Naive Bayes Classifier |
朴素贝叶斯分类器 |
| Named entity recognition |
命名实体识别 |
| Nash equilibrium |
纳什均衡 |
| nats |
奈特 |
| Natural language generation/NLG |
自然语言生成 |
| Natural language processing |
自然语言处理 |
| nearest neighbor |
最近邻 |
| nearest neighbor graph |
最近邻图 |
| nearest neighbor regression |
最近邻回归 |
| Negative class |
负类 |
| Negative correlation |
负相关法 |
| negative definite |
负定 |
| Negative Log Likelihood |
负对数似然 |
| negative part function |
负部函数 |
| negative phase |
负相 |
| negative semidefinite |
半负定 |
| Neighbourhood Component Analysis/NCA |
近邻成分分析 |
| Nesterovmomentum |
Nesterov 动量 |
| network |
网络 |
| neural auto-regressive densityestimator |
神经自回归密度估计器 |
| neural auto-regressive network |
神经自回归网络 |
| Neural Language Model |
神经语言模型 |
| Neural Machine Translation |
神经机器翻译 |
| neural network |
神经网络 |
| Neural Turing Machine |
神经图灵机/神经网络图灵机 |
| Newton method |
牛顿法 |
| n-gram |
n-gram |
| NIPS |
国际神经信息处理系统会议 |
| no free lunch theorem |
没有免费午餐定理 |
| No Free Lunch Theorem／NFL |
没有免费的午餐定理 |
| noise |
噪声 |
| noise distribution |
噪声分布 |
| Noise-contrastive estimation |
噪音对比估计 |
| Nominal attribute |
列名属性 |
| nonconvex |
非凸 |
| Non-convex optimization |
非凸优化 |
| nondistributed |
非分布式 |
| nondistributed representation |
非分布式表示 |
| nonlinear conjugate gradients |
非线性共轭梯度 |
| nonlinear independent componentsestimation |
非线性独立成分估计 |
| Nonlinear model |
非线性模型 |
| Non-metric distance |
非度量距离 |
| Non-negative matrix factorization |
非负矩阵分解 |
| Non-ordinal attribute |
无序属性 |
| non-parametric |
非参数 |
| Non-Saturating Game |
非饱和博弈 |
| Norm |
范数 |
| normal distribution |
正态分布 |
| normal equation |
正规方程 |
| Normalization |
归一化 |
| normalized |
归一化的 |
| normalized initialization |
标准初始化 |
| Nuclear norm |
核范数 |
| numeric value |
数值 |
| Numerical attribute |
数值属性 |
| numerical optimization |
数值优化 |
| object recognition |
对象识别 |
| objective |
目标 |
| Objective function |
目标函数 |
| Oblique decision tree |
斜决策树 |
| Occam’s razor |
奥卡姆剃刀 |
| Odds |
几率 |
| Off-Policy |
离策略 |
| One shot learning |
一次性学习 |
| One-Dependent Estimator／ODE |
独依赖估计 |
| one-hot |
one-hot |
| one-shot learning |
一次学习 |
| online |
在线 |
| online learning |
在线学习 |
| On-Policy |
在策略 |
| operation |
操作 |
| optimal capacity |
最佳容量 |
| Ordinal attribute |
有序属性 |
| origin |
原点 |
| orthogonal |
正交 |
| orthogonal matrix |
正交矩阵 |
| orthonormal |
标准正交 |
| Out-of-bag estimate |
包外估计 |
| output |
输出 |
| Output layer |
输出层 |
| Output smearing |
输出调制法 |
| overcomplete |
过完备 |
| overestimation |
过估计 |
| Overfitting |
过拟合／过配 |
| overfitting regime |
过拟合机制 |
| overflow |
上溢 |
| Oversampling |
过采样 |
| Paired t-test |
成对 t 检验 |
| Pairwise |
成对型 |
| Pairwise Markov property |
成对马尔可夫性 |
| Parallel Distributed Processing |
并行分布式处理 |
| parallel tempering |
并行回火 |
| Parameter |
参数 |
| Parameter estimation |
参数估计 |
| parameter server |
参数服务器 |
| parameter sharing |
参数共享 |
| Parameter tuning |
调参 |
| parametric case |
有参情况 |
| parametric ReLU |
参数化整流线性单元 |
| Parse tree |
解析树 |
| partial derivative |
偏导数 |
| Particle Swarm Optimization/PSO |
粒子群优化算法 |
| Partition Function |
配分函数 |
| Part-of-speech tagging |
词性标注 |
| PCA probabilistic PCA |
概率 |
| Perceptron |
感知机 |
| performance measures |
性能度量 |
| performance metrics |
性能度量 |
| permutation invariant |
置换不变性 |
| persistent contrastivedivergence |
持续性对比散度 |
| phoneme |
音素 |
| phonetic |
语音 |
| piecewise |
分段 |
| Plug and Play Generative Network |
即插即用生成网络 |
| Plurality voting |
相对多数投票法 |
| point estimator |
点估计 |
| Polarity detection |
极性检测 |
| policy |
策略 |
| policy gradient |
策略梯度 |
| Polynomial kernel function |
多项式核函数 |
| pool |
池 |
| Pooling |
池化 |
| pooling function |
池化函数 |
| poor conditioning |
病态条件 |
| Positive class |
正类 |
| positive definite |
正定 |
| Positive definite matrix |
正定矩阵 |
| positive part function |
正部函数 |
| positive phase |
正相 |
| positive semidefinite |
半正定 |
| posterior probability |
后验概率 |
| Post-hoc test |
后续检验 |
| Post-pruning |
后剪枝 |
| potential function |
势函数 |
| power method |
幂方法 |
| PR curve |
PR 曲线 |
| Precision |
查准率／准确率 |
| precision matrix |
精度矩阵 |
| predictive sparse decomposition |
预测稀疏分解 |
| Prepruning |
预剪枝 |
| pretraining |
预训练 |
| primary visual cortex |
初级视觉皮层 |
| Principal component analysis/PCA |
主成分分析 |
| Principle of multiple explanations |
多释原则 |
| Prior |
先验 |
| prior probability |
先验概率 |
| prior probability distribution |
先验概率分布 |
| probability density function |
概率密度函数 |
| probability distribution |
概率分布 |
| Probability Graphical Model |
概率图模型 |
| probability mass function |
概率质量函数 |
| product of expert |
专家之积 |
| product rule |
乘法法则 |
| proportional |
成比例 |
| proposal distribution |
提议分布 |
| Proximal Gradient Descent/PGD |
近端梯度下降 |
| Pruning |
剪枝 |
| Pseudo-label |
伪标记 |
| pseudolikelihood |
伪似然 |
| quadrature pair |
象限对 |
| Quantized Neural Network |
量子化神经网络 |
| Quantum computer |
量子计算机 |
| Quantum Computing |
量子计算 |
| quantum mechanics |
量子力学 |
| Quasi Newton method |
拟牛顿法 |
| Radial Basis Function／RBF |
径向基函数 |
| Random Forest Algorithm |
随机森林算法 |
| random search |
随机搜索 |
| random variable |
随机变量 |
| Random walk |
随机漫步 |
| range |
值域 |
| ratio matching |
比率匹配 |
| RBM discriminative RBM |
判别 |
| RBM Gaussian RBM |
高斯 |
| Recall |
查全率／召回率 |
| Receiver Operating Characteristic/ROC |
受试者工作特征 |
| receptive field |
接受域 |
| recirculation |
再循环 |
| recommender system |
推荐系统 |
| reconstruction |
重构 |
| reconstruction error |
重构误差 |
| rectified linear |
整流线性 |
| rectified linear transformation |
整流线性变换 |
| rectified linear unit |
整流线性单元 |
| Rectified Linear Unit/ReLU |
线性修正单元 |
| rectifier network |
整流网络 |
| recurrence |
循环 |
| recurrent convolutional network |
循环卷积网络 |
| recurrent network |
循环网络 |
| Recurrent Neural Network |
循环神经网络 |
| Recursive neural network |
递归神经网络 |
| Reference model |
参考模型 |
| Regression |
回归 |
| Regularization |
正则化 |
| regularize |
正则化 |
| regularizer |
正则化项 |
| Reinforcement learning/RL |
强化学习 |
| relation |
关系 |
| relational database |
关系型数据库 |
| reparametrization |
重参数化 |
| reparametrization trick |
重参数化技巧 |
| representation |
表示 |
| Representation learning |
表征学习/表示学习 |
| representational capacity |
表示容量 |
| Representer theorem |
表示定理 |
| reproducing kernel Hilbert space/RKHS |
再生核希尔伯特空间 |
| Re-sampling |
重采样法 |
| Rescaling |
再缩放 |
| reservoir computing |
储层计算 |
| Residual Mapping |
残差映射 |
| Residual Network |
残差网络 |
| Restricted Boltzmann Machine/RBM |
受限玻尔兹曼机 |
| Restricted Isometry Property/RIP |
限定等距性 |
| reverse correlation |
反向相关 |
| reverse mode accumulation |
反向模式累加 |
| Re-weighting |
重赋权法 |
| ridge regression |
岭回归 |
| right eigenvector |
右特征向量 |
| right singular vector |
右奇异向量 |
| risk |
风险 |
| RNN gated RNN |
门控 |
| Robustness |
稳健性/鲁棒性 |
| Root node |
根结点 |
| row |
行 |
| Rule Engine |
规则引擎 |
| Rule learning |
规则学习 |
| saccade |
扫视 |
| Saddle point |
鞍点 |
| saddle-free Newton method |
无鞍牛顿法 |
| same |
相同 |
| sample mean |
样本均值 |
| Sample space |
样本空间 |
| sample variance |
样本方差 |
| Sampling |
采样 |
| saturate |
饱和 |
| scalar |
标量 |
| score |
得分 |
| Score function |
评分函数 |
| score matching |
得分匹配 |
| second derivative |
二阶导数 |
| second derivative test |
二阶导数测试 |
| second layer |
第二层 |
| second-order method |
二阶方法 |
| self-contrastive estimation |
自对比估计 |
| Self-Driving |
自动驾驶 |
| self-information |
自信息 |
| Self-Organizing Map／SOM |
自组织映射 |
| semantic hashing |
语义哈希 |
| Semi-naive Bayes classifiers |
半朴素贝叶斯分类器 |
| semi-restricted Boltzmann Machine |
半受限玻尔兹曼机 |
| semi-supervised |
半监督 |
| Semi-Supervised Learning |
半监督学习 |
| semi-Supervised Support Vector Machine |
半监督支持向量机 |
| Sentiment analysis |
情感分析 |
| separable |
可分离的 |
| separate |
分离的 |
| Separating hyperplane |
分离超平面 |
| separation |
分离 |
| setting |
情景 |
| shadow circuit |
浅度回路 |
| Shannon entropy |
香农熵 |
| shannons |
香农 |
| shaping |
塑造 |
| shortlist |
短列表 |
| sigmoid |
sigmoid |
| Sigmoid function |
Sigmoid 函数 |
| sigmoidBelief Network |
sigmoid 信念网络 |
| Similarity measure |
相似度度量 |
| simple cell |
简单细胞 |
| Simulated annealing |
模拟退火 |
| Simultaneous localization and mapping |
同步定位与地图构建 |
| singular |
奇异的 |
| singular value |
奇异值 |
| Singular Value Decomposition |
奇异值分解 |
| singular vector |
奇异向量 |
| skip connection |
跳跃连接 |
| Slack variables |
松弛变量 |
| slow feature analysis |
慢特征分析 |
| slowness principle |
慢性原则 |
| Smoothing |
平滑 |
| smoothness prior |
平滑先验 |
| Soft margin |
软间隔 |
| Soft margin maximization |
软间隔最大化 |
| Soft voting |
软投票 |
| softmax |
softmax |
| softmaxfunction |
softmax 函数 |
| softmaxunit |
softmax 单元 |
| softplus |
softplus |
| softplusfunction |
softplus 函数 |
| span |
生成子空间 |
| sparse |
稀疏 |
| sparse activation |
稀疏激活 |
| sparse coding |
稀疏编码 |
| sparse connectivity |
稀疏连接 |
| sparse initialization |
稀疏初始化 |
| sparse interactions |
稀疏交互 |
| Sparse representation |
稀疏表征 |
| sparse weights |
稀疏权重 |
| Sparsity |
稀疏性 |
| Specialization |
特化 |
| Spectral Clustering |
谱聚类 |
| spectral radius |
谱半径 |
| Speech Recognition |
语音识别 |
| sphering |
sphering |
| spike and slab |
尖峰和平板 |
| spike and slab RBM |
尖峰和平板RBM |
| Splitting variable |
切分变量 |
| spurious modes |
虚假模态 |
| square |
方阵 |
| Squashing function |
挤压函数 |
| Stability-plasticity dilemma |
可塑性-稳定性困境 |
| standard deviation |
标准差 |
| standard error |
标准差 |
| standard normal distribution |
标准正态分布 |
| statement |
声明 |
| stationary |
平稳的 |
| Stationary Distribution |
平稳分布 |
| stationary point |
驻点 |
| statistic efficiency |
统计效率 |
| Statistical learning |
统计学习 |
| statistical learning theory |
统计学习理论 |
| statistics |
统计量 |
| Status feature function |
状态特征函 |
| steepest descent |
最陡下降 |
| stochastic |
随机 |
| stochastic curriculum |
随机课程 |
| Stochastic Gradient Ascent |
随机梯度上升 |
| Stochastic gradient descent |
随机梯度下降 |
| Stochastic Matrix |
随机矩阵 |
| stochastic maximum likelihood |
随机最大似然 |
| Stratified sampling |
分层采样 |
| stream |
流 |
| stride |
步幅 |
| Structural risk |
结构风险 |
| Structural risk minimization/SRM |
结构风险最小化 |
| structure learning |
结构学习 |
| structured probabilistic model |
结构化概率模型 |
| structured variational inference |
结构化变分推断 |
| subatomic |
亚原子 |
| subsample |
子采样 |
| Subspace |
子空间 |
| sum rule |
求和法则 |
| sum-productnetwork |
和-积网络 |
| supervised |
监督 |
| Supervised learning |
监督学习／有导师学习 |
| supervised learning algorithm |
监督学习算法 |
| supervised model |
监督模型 |
| supervised pretraining |
监督预训练 |
| support vector |
支持向量 |
| support vector expansion |
支持向量展式 |
| Support Vector Machine/SVM |
支持向量机 |
| Surrogat loss |
替代损失 |
| Surrogate function |
替代函数 |
| surrogate loss function |
代理损失函数 |
| symbol |
符号 |
| Symbolic learning |
符号学习 |
| symbolic representation |
符号表示 |
| Symbolism |
符号主义 |
| symmetric |
对称 |
| Synset |
同义词集 |
| tangent distance |
切面距离 |
| tangent plane |
切平面 |
| tangent prop |
正切传播 |
| tangent propagation |
正切传播 |
| target |
目标 |
| taylor |
泰勒 |
| T-Distribution Stochastic Neighbour Embedding/t-SNE |
T – 分布随机近邻嵌入 |
| teacher forcing |
导师驱动过程 |
| temperature |
温度 |
| tempered transition |
回火转移 |
| tempering |
回火 |
| Tensor |
张量 |
| Tensor Processing Units/TPU |
张量处理单元 |
| test error |
测试误差 |
| test set |
测试集 |
| the collider case |
碰撞情况 |
| The least square method |
最小二乘法 |
| Threshold |
阈值 |
| Threshold logic unit |
阈值逻辑单元 |
| Threshold-moving |
阈值移动 |
| tied weights |
绑定的权重 |
| Tikhonovregularization |
Tikhonov 正则 |
| tiled convolution |
平铺卷积 |
| time delay neural network |
时延神经网络 |
| Time Step |
时间步骤 |
| Toeplitzmatrix |
Toeplitz 矩阵 |
| token |
标记 |
| Tokenization |
标记化 |
| tolerance |
容差 |
| Training error |
训练误差 |
| Training instance |
训练示例／训练例 |
| training set |
训练集 |
| transcribe |
转录 |
| transcription system |
转录系统 |
| Transductive learning |
直推学习 |
| Transfer learning |
迁移学习 |
| transition |
转移 |
| transpose |
转置 |
| Treebank |
树库 |
| Tria-by-error |
试错法 |
| triangle inequality |
三角不等式 |
| triangulate |
三角形化 |
| triangulated graph |
三角形化图 |
| trigram |
三元语法 |
| True negative |
真负类 |
| True positive |
真正类 |
| True Positive Rate/TPR |
真正例率 |
| Turing Machine |
图灵机 |
| Twice-learning |
二次学习 |
| unbiased |
无偏 |
| unbiased sample variance |
无偏样本方差 |
| undercomplete |
欠完备 |
| underdetermined |
欠定的 |
| underestimation |
欠估计 |
| Underfitting |
欠拟合／欠配 |
| underfitting regime |
欠拟合机制 |
| underflow |
下溢 |
| underlying |
潜在 |
| underlying cause |
潜在成因 |
| Undersampling |
欠采样 |
| Understandability |
可理解性 |
| undirected |
无向 |
| undirected graphical model |
无向图模型 |
| undirected Model |
无向模型 |
| Unequal cost |
非均等代价 |
| unfolded graph |
展开图 |
| unfolding |
展开 |
| uniform distribution |
均匀分布 |
| unigram |
一元语法 |
| unimodal |
单峰值 |
| unit |
单元 |
| unit norm |
单位范数 |
| unit vector |
单位向量 |
| Unit-step function |
单位阶跃函数 |
| Univariate decision tree |
单变量决策树 |
| universal approximation theorem |
万能近似定理 |
| universal approximator |
万能近似器 |
| universal function approximator |
万能函数近似器 |
| unlabeled |
未标注 |
| unnormalized probabilityfunction |
未归一化概率函数 |
| unshared convolution |
非共享卷积 |
| unsupervised |
无监督 |
| Unsupervised layer-wise training |
无监督逐层训练 |
| Unsupervised learning |
无监督学习／无导师学习 |
| unsupervised learning algorithm |
无监督学习算法 |
| unsupervised pretraining |
无监督预训练 |
| Upsampling |
上采样 |
| valid |
有效 |
| validation set |
验证集 |
| vanishing and exploding gradientproblem |
梯度消失与爆炸问题 |
| vanishing gradient |
梯度消失 |
| Vanishing Gradient Problem |
梯度消失问题 |
| Vapnik-Chervonenkis dimension |
Vapnik-Chervonenkis 维度 |
| variable elimination |
变量消去 |
| variance |
方差 |
| variance reduction |
方差减小 |
| variational auto-encoder |
变分自编码器 |
| variational derivative |
变分导数 |
| variational free energy |
变分自由能 |
| Variational inference |
变分推断 |
| VC Theory |
VC维理论 |
| vector |
向量 |
| Version space |
版本空间 |
| virtual adversarial example |
虚拟对抗样本 |
| virtual adversarial training |
虚拟对抗训练 |
| visible layer |
可见层 |
| Viterbi algorithm |
维特比算法 |
| Von Neumann architecture |
冯 · 诺伊曼架构 |
| V-structure |
V-结构 |
| wake sleep |
醒眠 |
| warp |
warp |
| Wasserstein GAN/WGAN |
Wasserstein生成对抗网络 |
| Weak learner |
弱学习器 |
| Weight |
权重 |
| weight decay |
权重衰减 |
| weight scaling inference rule |
权重比例推断规则 |
| Weight sharing |
权共享 |
| weight space symmetry |
权重空间对称性 |
| Weighted voting |
加权投票法 |
| whitening |
白化 |
| width |
宽度 |
| winner-take-all |
赢者通吃 |
| Within-class scatter matrix |
类内散度矩阵 |
| Word embedding |
词嵌入 |
| word-sense disambiguation |
词义消歧 |
| Zero-data learning |
零数据学习 |
| Zero-shot learning |
零次学习 |
