下面几位机器学习权威专家汇总的725个机器学习术语表，非常全面了，值得收藏！
<style type="text/css">
th{padding:5px;}
td{padding:5px;}
</style>
<table border="1" bgcolor="#ececec">
<thead>
<tr><th>英文术语</th><th>中文翻译</th></tr>
</thead>
<tbody>
<tr><td>0-1 Loss Function</td><td>0-1损失函数</td>
</tr>
<tr><td>Accept-Reject Sampling Method</td><td>接受-拒绝抽样法/接受-拒绝采样法</td>
</tr>
<tr><td>Accumulated Error Backpropagation</td><td>累积误差反向传播</td>
</tr>
<tr><td>Accuracy</td><td>精度</td>
</tr>
<tr><td>Acquisition Function</td><td>采集函数</td>
</tr>
<tr><td>Action</td><td>动作</td>
</tr>
<tr><td>Activation Function</td><td>激活函数</td>
</tr>
<tr><td>Active Learning</td><td>主动学习</td>
</tr>
<tr><td>Adaptive Bitrate Algorithm</td><td>自适应比特率算法</td>
</tr>
<tr><td>Adaptive Boosting</td><td>AdaBoost</td>
</tr>
<tr><td>Adaptive Gradient Algorithm</td><td>AdaGrad</td>
</tr>
<tr><td>Adaptive Moment Estimation Algorithm</td><td>Adam算法</td>
</tr>
<tr><td>Adaptive Resonance Theory</td><td>自适应谐振理论</td>
</tr>
<tr><td>Additive Model</td><td>加性模型</td>
</tr>
<tr><td>Affinity Matrix</td><td>亲和矩阵</td>
</tr>
<tr><td>Agent</td><td>智能体</td>
</tr>
<tr><td>Algorithm</td><td>算法</td>
</tr>
<tr><td>Alpha-Beta Pruning</td><td>α-β修剪法</td>
</tr>
<tr><td>Anomaly Detection</td><td>异常检测</td>
</tr>
<tr><td>Approximate Inference</td><td>近似推断</td>
</tr>
<tr><td>Area Under ROC Curve</td><td>AUC</td>
</tr>
<tr><td>Artificial Intelligence</td><td>人工智能</td>
</tr>
<tr><td>Artificial Neural Network</td><td>人工神经网络</td>
</tr>
<tr><td>Artificial Neuron</td><td>人工神经元</td>
</tr>
<tr><td>Attention</td><td>注意力</td>
</tr>
<tr><td>Attention Mechanism</td><td>注意力机制</td>
</tr>
<tr><td>Attribute</td><td>属性</td>
</tr>
<tr><td>Attribute Space</td><td>属性空间</td>
</tr>
<tr><td>Autoencoder</td><td>自编码器</td>
</tr>
<tr><td>Automatic Differentiation</td><td>自动微分</td>
</tr>
<tr><td>Autoregressive Model</td><td>自回归模型</td>
</tr>
<tr><td>Back Propagation</td><td>反向传播</td>
</tr>
<tr><td>Back Propagation Algorithm</td><td>反向传播算法</td>
</tr>
<tr><td>Back Propagation Through Time</td><td>随时间反向传播</td>
</tr>
<tr><td>Backward Induction</td><td>反向归纳</td>
</tr>
<tr><td>Backward Search</td><td>反向搜索</td>
</tr>
<tr><td>Bag of Words</td><td>词袋</td>
</tr>
<tr><td>Bandit</td><td>赌博机/老虎机</td>
</tr>
<tr><td>Base Learner</td><td>基学习器</td>
</tr>
<tr><td>Base Learning Algorithm</td><td>基学习算法</td>
</tr>
<tr><td>Baseline</td><td>基准</td>
</tr>
<tr><td>Batch</td><td>批量</td>
</tr>
<tr><td>Batch Normalization</td><td>批量规范化</td>
</tr>
<tr><td>Bayes Decision Rule</td><td>贝叶斯决策准则</td>
</tr>
<tr><td>Bayes Model Averaging</td><td>贝叶斯模型平均</td>
</tr>
<tr><td>Bayes Optimal Classifier</td><td>贝叶斯最优分类器</td>
</tr>
<tr><td>Bayes' Theorem</td><td>贝叶斯定理</td>
</tr>
<tr><td>Bayesian Decision Theory</td><td>贝叶斯决策理论</td>
</tr>
<tr><td>Bayesian Inference</td><td>贝叶斯推断</td>
</tr>
<tr><td>Bayesian Learning</td><td>贝叶斯学习</td>
</tr>
<tr><td>Bayesian Network</td><td>贝叶斯网/贝叶斯网络</td>
</tr>
<tr><td>Bayesian Optimization</td><td>贝叶斯优化</td>
</tr>
<tr><td>Beam Search</td><td>束搜索</td>
</tr>
<tr><td>Benchmark</td><td>基准</td>
</tr>
<tr><td>Belief Network</td><td>信念网/信念网络</td>
</tr>
<tr><td>Belief Propagation</td><td>信念传播</td>
</tr>
<tr><td>Bellman Equation</td><td>贝尔曼方程</td>
</tr>
<tr><td>Bernoulli Distribution</td><td>伯努利分布</td>
</tr>
<tr><td>Beta Distribution</td><td>贝塔分布</td>
</tr>
<tr><td>Between-Class Scatter Matrix</td><td>类间散度矩阵</td>
</tr>
<tr><td>BFGS</td><td>BFGS</td>
</tr>
<tr><td>Bias</td><td>偏差/偏置</td>
</tr>
<tr><td>Bias In Affine Function</td><td>偏置</td>
</tr>
<tr><td>Bias In Statistics</td><td>偏差</td>
</tr>
<tr><td>Bias Shift</td><td>偏置偏移</td>
</tr>
<tr><td>Bias-Variance Decomposition</td><td>偏差 - 方差分解</td>
</tr>
<tr><td>Bias-Variance Dilemma</td><td>偏差 - 方差困境</td>
</tr>
<tr><td>Bidirectional Recurrent Neural Network</td><td>双向循环神经网络</td>
</tr>
<tr><td>Bigram</td><td>二元语法</td>
</tr>
<tr><td>Bilingual Evaluation Understudy</td><td>BLEU</td>
</tr>
<tr><td>Binary Classification</td><td>二分类</td>
</tr>
<tr><td>Binomial Distribution</td><td>二项分布</td>
</tr>
<tr><td>Binomial Test</td><td>二项检验</td>
</tr>
<tr><td>Boltzmann Distribution</td><td>玻尔兹曼分布</td>
</tr>
<tr><td>Boltzmann Machine</td><td>玻尔兹曼机</td>
</tr>
<tr><td>Boosting</td><td>Boosting</td>
</tr>
<tr><td>Bootstrap Aggregating</td><td>Bagging</td>
</tr>
<tr><td>Bootstrap Sampling</td><td>自助采样法</td>
</tr>
<tr><td>Bootstrapping</td><td>自助法/自举法</td>
</tr>
<tr><td>Break-Event Point</td><td>平衡点</td>
</tr>
<tr><td>Bucketing</td><td>分桶</td>
</tr>
<tr><td>Calculus of Variations</td><td>变分法</td>
</tr>
<tr><td>Cascade-Correlation</td><td>级联相关</td>
</tr>
<tr><td>Catastrophic Forgetting</td><td>灾难性遗忘</td>
</tr>
<tr><td>Categorical Distribution</td><td>类别分布</td>
</tr>
<tr><td>Cell</td><td>单元</td>
</tr>
<tr><td>Chain Rule</td><td>链式法则</td>
</tr>
<tr><td>Chebyshev Distance</td><td>切比雪夫距离</td>
</tr>
<tr><td>Class</td><td>类别</td>
</tr>
<tr><td>Class-Imbalance</td><td>类别不平衡</td>
</tr>
<tr><td>Classification</td><td>分类</td>
</tr>
<tr><td>Classification And Regression Tree</td><td>分类与回归树</td>
</tr>
<tr><td>Classifier</td><td>分类器</td>
</tr>
<tr><td>Clique</td><td>团</td>
</tr>
<tr><td>Cluster</td><td>簇</td>
</tr>
<tr><td>Cluster Assumption</td><td>聚类假设</td>
</tr>
<tr><td>Clustering</td><td>聚类</td>
</tr>
<tr><td>Clustering Ensemble</td><td>聚类集成</td>
</tr>
<tr><td>Co-Training</td><td>协同训练</td>
</tr>
<tr><td>Coding Matrix</td><td>编码矩阵</td>
</tr>
<tr><td>Collaborative Filtering</td><td>协同过滤</td>
</tr>
<tr><td>Competitive Learning</td><td>竞争型学习</td>
</tr>
<tr><td>Comprehensibility</td><td>可解释性</td>
</tr>
<tr><td>Computation Graph</td><td>计算图</td>
</tr>
<tr><td>Computational Learning Theory</td><td>计算学习理论</td>
</tr>
<tr><td>Conditional Entropy</td><td>条件熵</td>
</tr>
<tr><td>Conditional Probability</td><td>条件概率</td>
</tr>
<tr><td>Conditional Probability Distribution</td><td>条件概率分布</td>
</tr>
<tr><td>Conditional Random Field</td><td>条件随机场</td>
</tr>
<tr><td>Conditional Risk</td><td>条件风险</td>
</tr>
<tr><td>Confidence</td><td>置信度</td>
</tr>
<tr><td>Confusion Matrix</td><td>混淆矩阵</td>
</tr>
<tr><td>Conjugate Distribution</td><td>共轭分布</td>
</tr>
<tr><td>Connection Weight</td><td>连接权</td>
</tr>
<tr><td>Connectionism</td><td>连接主义</td>
</tr>
<tr><td>Consistency</td><td>一致性</td>
</tr>
<tr><td>Constrained Optimization</td><td>约束优化</td>
</tr>
<tr><td>Context Variable</td><td>上下文变量</td>
</tr>
<tr><td>Context Vector</td><td>上下文向量</td>
</tr>
<tr><td>Context Window</td><td>上下文窗口</td>
</tr>
<tr><td>Context Word</td><td>上下文词</td>
</tr>
<tr><td>Contextual Bandit</td><td>上下文赌博机/上下文老虎机</td>
</tr>
<tr><td>Contingency Table</td><td>列联表</td>
</tr>
<tr><td>Continuous Attribute</td><td>连续属性</td>
</tr>
<tr><td>Contrastive Divergence</td><td>对比散度</td>
</tr>
<tr><td>Convergence</td><td>收敛</td>
</tr>
<tr><td>Convex Optimization</td><td>凸优化</td>
</tr>
<tr><td>Convex Quadratic Programming</td><td>凸二次规划</td>
</tr>
<tr><td>Convolution</td><td>卷积</td>
</tr>
<tr><td>Convolutional Kernel</td><td>卷积核</td>
</tr>
<tr><td>Convolutional Neural Network</td><td>卷积神经网络</td>
</tr>
<tr><td>Coordinate Descent</td><td>坐标下降</td>
</tr>
<tr><td>Corpus</td><td>语料库</td>
</tr>
<tr><td>Correlation Coefficient</td><td>相关系数</td>
</tr>
<tr><td>Cosine Similarity</td><td>余弦相似度</td>
</tr>
<tr><td>Cost</td><td>代价</td>
</tr>
<tr><td>Cost Curve</td><td>代价曲线</td>
</tr>
<tr><td>Cost Function</td><td>代价函数</td>
</tr>
<tr><td>Cost Matrix</td><td>代价矩阵</td>
</tr>
<tr><td>Cost-Sensitive</td><td>代价敏感</td>
</tr>
<tr><td>Covariance</td><td>协方差</td>
</tr>
<tr><td>Covariance Matrix</td><td>协方差矩阵</td>
</tr>
<tr><td>Critical Point</td><td>临界点</td>
</tr>
<tr><td>Cross Entropy</td><td>交叉熵</td>
</tr>
<tr><td>Cross Validation</td><td>交叉验证</td>
</tr>
<tr><td>Curse of Dimensionality</td><td>维数灾难</td>
</tr>
<tr><td>Cutting Plane Algorithm</td><td>割平面法</td>
</tr>
<tr><td>Data Mining</td><td>数据挖掘</td>
</tr>
<tr><td>Data Set</td><td>数据集</td>
</tr>
<tr><td>Davidon-Fletcher-Powell</td><td>DFP</td>
</tr>
<tr><td>Decision Boundary</td><td>决策边界</td>
</tr>
<tr><td>Decision Function</td><td>决策函数</td>
</tr>
<tr><td>Decision Stump</td><td>决策树桩</td>
</tr>
<tr><td>Decision Tree</td><td>决策树</td>
</tr>
<tr><td>Decoder</td><td>解码器</td>
</tr>
<tr><td>Decoding</td><td>解码</td>
</tr>
<tr><td>Deconvolution</td><td>反卷积</td>
</tr>
<tr><td>Deconvolutional Network</td><td>反卷积网络</td>
</tr>
<tr><td>Deduction</td><td>演绎</td>
</tr>
<tr><td>Deep Belief Network</td><td>深度信念网络</td>
</tr>
<tr><td>Deep Boltzmann Machine</td><td>深度玻尔兹曼机</td>
</tr>
<tr><td>Deep Convolutional Generative Adversarial Network</td><td>深度卷积生成对抗网络</td>
</tr>
<tr><td>Deep Learning</td><td>深度学习</td>
</tr>
<tr><td>Deep Neural Network</td><td>深度神经网络</td>
</tr>
<tr><td>Deep Q-Network</td><td>深度Q网络</td>
</tr>
<tr><td>Delta-Bar-Delta</td><td>Delta-Bar-Delta</td>
</tr>
<tr><td>Denoising</td><td>去噪</td>
</tr>
<tr><td>Denoising Autoencoder</td><td>去噪自编码器</td>
</tr>
<tr><td>Denoising Score Matching</td><td>去躁分数匹配</td>
</tr>
<tr><td>Density Estimation</td><td>密度估计</td>
</tr>
<tr><td>Density-Based Clustering</td><td>密度聚类</td>
</tr>
<tr><td>Derivative</td><td>导数</td>
</tr>
<tr><td>Determinant</td><td>行列式</td>
</tr>
<tr><td>Diagonal Matrix</td><td>对角矩阵</td>
</tr>
<tr><td>Dictionary Learning</td><td>字典学习</td>
</tr>
<tr><td>Dimension Reduction</td><td>降维</td>
</tr>
<tr><td>Directed Edge</td><td>有向边</td>
</tr>
<tr><td>Directed Graphical Model</td><td>有向图模型</td>
</tr>
<tr><td>Directed Separation</td><td>有向分离</td>
</tr>
<tr><td>Dirichlet Distribution</td><td>狄利克雷分布</td>
</tr>
<tr><td>Discriminative Model</td><td>判别式模型</td>
</tr>
<tr><td>Discriminator</td><td>判别器</td>
</tr>
<tr><td>Discriminator Network</td><td>判别网络</td>
</tr>
<tr><td>Distance Measure</td><td>距离度量</td>
</tr>
<tr><td>Distance Metric Learning</td><td>距离度量学习</td>
</tr>
<tr><td>Distributed Representation</td><td>分布式表示</td>
</tr>
<tr><td>Diverge</td><td>发散</td>
</tr>
<tr><td>Divergence</td><td>散度</td>
</tr>
<tr><td>Diversity</td><td>多样性</td>
</tr>
<tr><td>Diversity Measure</td><td>多样性度量/差异性度量</td>
</tr>
<tr><td>Domain Adaptation</td><td>领域自适应</td>
</tr>
<tr><td>Dominant Strategy</td><td>主特征值</td>
</tr>
<tr><td>Dominant Strategy</td><td>占优策略</td>
</tr>
<tr><td>Down Sampling</td><td>下采样</td>
</tr>
<tr><td>Dropout</td><td>暂退法</td>
</tr>
<tr><td>Dropout Boosting</td><td>暂退Boosting</td>
</tr>
<tr><td>Dropout Method</td><td>暂退法</td>
</tr>
<tr><td>Dual Problem</td><td>对偶问题</td>
</tr>
<tr><td>Dummy Node</td><td>哑结点</td>
</tr>
<tr><td>Dynamic Bayesian Network</td><td>动态贝叶斯网络</td>
</tr>
<tr><td>Dynamic Programming</td><td>动态规划</td>
</tr>
<tr><td>Early Stopping</td><td>早停</td>
</tr>
<tr><td>Eigendecomposition</td><td>特征分解</td>
</tr>
<tr><td>Eigenvalue</td><td>特征值</td>
</tr>
<tr><td>Element-Wise Product</td><td>逐元素积</td>
</tr>
<tr><td>Embedding</td><td>嵌入</td>
</tr>
<tr><td>Empirical Conditional Entropy</td><td>经验条件熵</td>
</tr>
<tr><td>Empirical Distribution</td><td>经验分布</td>
</tr>
<tr><td>Empirical Entropy</td><td>经验熵</td>
</tr>
<tr><td>Empirical Error</td><td>经验误差</td>
</tr>
<tr><td>Empirical Risk</td><td>经验风险</td>
</tr>
<tr><td>Empirical Risk Minimization</td><td>经验风险最小化</td>
</tr>
<tr><td>Encoder</td><td>编码器</td>
</tr>
<tr><td>Encoding</td><td>编码</td>
</tr>
<tr><td>End-To-End</td><td>端到端</td>
</tr>
<tr><td>Energy Function</td><td>能量函数</td>
</tr>
<tr><td>Energy-Based Model</td><td>基于能量的模型</td>
</tr>
<tr><td>Ensemble Learning</td><td>集成学习</td>
</tr>
<tr><td>Ensemble Pruning</td><td>集成修剪</td>
</tr>
<tr><td>Entropy</td><td>熵</td>
</tr>
<tr><td>Episode</td><td>回合</td>
</tr>
<tr><td>Epoch</td><td>轮</td>
</tr>
<tr><td>Error</td><td>误差</td>
</tr>
<tr><td>Error Backpropagation Algorithm</td><td>误差反向传播算法</td>
</tr>
<tr><td>Error Backpropagation</td><td>误差反向传播</td>
</tr>
<tr><td>Error Correcting Output Codes</td><td>纠错输出编码</td>
</tr>
<tr><td>Error Rate</td><td>错误率</td>
</tr>
<tr><td>Error-Ambiguity Decomposition</td><td>误差－分歧分解</td>
</tr>
<tr><td>Estimator</td><td>估计/估计量</td>
</tr>
<tr><td>Euclidean Distance</td><td>欧氏距离</td>
</tr>
<tr><td>Evidence</td><td>证据</td>
</tr>
<tr><td>Evidence Lower Bound</td><td>证据下界</td>
</tr>
<tr><td>Exact Inference</td><td>精确推断</td>
</tr>
<tr><td>Example</td><td>样例</td>
</tr>
<tr><td>Expectation</td><td>期望</td>
</tr>
<tr><td>Expectation Maximization</td><td>期望最大化</td>
</tr>
<tr><td>Expected Loss</td><td>期望损失</td>
</tr>
<tr><td>Expert System</td><td>专家系统</td>
</tr>
<tr><td>Exploding Gradient</td><td>梯度爆炸</td>
</tr>
<tr><td>Exponential Loss Function</td><td>指数损失函数</td>
</tr>
<tr><td>Factor</td><td>因子</td>
</tr>
<tr><td>Factorization</td><td>因子分解</td>
</tr>
<tr><td>Feature</td><td>特征</td>
</tr>
<tr><td>Feature Engineering</td><td>特征工程</td>
</tr>
<tr><td>Feature Map</td><td>特征图</td>
</tr>
<tr><td>Feature Selection</td><td>特征选择</td>
</tr>
<tr><td>Feature Vector</td><td>特征向量</td>
</tr>
<tr><td>Featured Learning</td><td>特征学习</td>
</tr>
<tr><td>Feedforward</td><td>前馈</td>
</tr>
<tr><td>Feedforward Neural Network</td><td>前馈神经网络</td>
</tr>
<tr><td>Few-Shot Learning</td><td>少试学习</td>
</tr>
<tr><td>Filter</td><td>滤波器</td>
</tr>
<tr><td>Fine-Tuning</td><td>微调</td>
</tr>
<tr><td>Fluctuation</td><td>振荡</td>
</tr>
<tr><td>Forget Gate</td><td>遗忘门</td>
</tr>
<tr><td>Forward Propagation</td><td>前向传播/正向传播</td>
</tr>
<tr><td>Forward Stagewise Algorithm</td><td>前向分步算法</td>
</tr>
<tr><td>Fractionally Strided Convolution</td><td>微步卷积</td>
</tr>
<tr><td>Frobenius Norm</td><td>Frobenius 范数</td>
</tr>
<tr><td>Full Padding</td><td>全填充</td>
</tr>
<tr><td>Functional</td><td>泛函</td>
</tr>
<tr><td>Functional Neuron</td><td>功能神经元</td>
</tr>
<tr><td>Gated Recurrent Unit</td><td>门控循环单元</td>
</tr>
<tr><td>Gated RNN</td><td>门控RNN</td>
</tr>
<tr><td>Gaussian Distribution</td><td>高斯分布</td>
</tr>
<tr><td>Gaussian Kernel</td><td>高斯核</td>
</tr>
<tr><td>Gaussian Kernel Function</td><td>高斯核函数</td>
</tr>
<tr><td>Gaussian Mixture Model</td><td>高斯混合模型</td>
</tr>
<tr><td>Gaussian Process</td><td>高斯过程</td>
</tr>
<tr><td>Generalization Ability</td><td>泛化能力</td>
</tr>
<tr><td>Generalization Error</td><td>泛化误差</td>
</tr>
<tr><td>Generalization Error Bound</td><td>泛化误差上界</td>
</tr>
<tr><td>Generalize</td><td>泛化</td>
</tr>
<tr><td>Generalized Lagrange Function</td><td>广义拉格朗日函数</td>
</tr>
<tr><td>Generalized Linear Model</td><td>广义线性模型</td>
</tr>
<tr><td>Generalized Rayleigh Quotient</td><td>广义瑞利商</td>
</tr>
<tr><td>Generative Adversarial Network</td><td>生成对抗网络</td>
</tr>
<tr><td>Generative Model</td><td>生成式模型</td>
</tr>
<tr><td>Generator</td><td>生成器</td>
</tr>
<tr><td>Generator Network</td><td>生成器网络</td>
</tr>
<tr><td>Genetic Algorithm</td><td>遗传算法</td>
</tr>
<tr><td>Gibbs Distribution</td><td>吉布斯分布</td>
</tr>
<tr><td>Gibbs Sampling</td><td>吉布斯采样/吉布斯抽样</td>
</tr>
<tr><td>Gini Index</td><td>基尼指数</td>
</tr>
<tr><td>Global Markov Property</td><td>全局马尔可夫性</td>
</tr>
<tr><td>Global Minimum</td><td>全局最小</td>
</tr>
<tr><td>Gradient</td><td>梯度</td>
</tr>
<tr><td>Gradient Clipping</td><td>梯度截断</td>
</tr>
<tr><td>Gradient Descent</td><td>梯度下降</td>
</tr>
<tr><td>Gradient Descent Method</td><td>梯度下降法</td>
</tr>
<tr><td>Gradient Exploding Problem</td><td>梯度爆炸问题</td>
</tr>
<tr><td>Gram Matrix</td><td>Gram 矩阵</td>
</tr>
<tr><td>Graph Convolutional Network</td><td>图卷积神经网络/图卷积网络</td>
</tr>
<tr><td>Graph Neural Network</td><td>图神经网络</td>
</tr>
<tr><td>Graphical Model</td><td>图模型</td>
</tr>
<tr><td>Grid Search</td><td>网格搜索</td>
</tr>
<tr><td>Ground Truth</td><td>真实值</td>
</tr>
<tr><td>Hadamard Product</td><td>Hadamard积</td>
</tr>
<tr><td>Hamming Distance</td><td>汉明距离</td>
</tr>
<tr><td>Hard Margin</td><td>硬间隔</td>
</tr>
<tr><td>Hebbian Rule</td><td>赫布法则</td>
</tr>
<tr><td>Hidden Layer</td><td>隐藏层</td>
</tr>
<tr><td>Hidden Markov Model</td><td>隐马尔可夫模型</td>
</tr>
<tr><td>Hidden Variable</td><td>隐变量</td>
</tr>
<tr><td>Hierarchical Clustering</td><td>层次聚类</td>
</tr>
<tr><td>Hilbert Space</td><td>希尔伯特空间</td>
</tr>
<tr><td>Hinge Loss Function</td><td>合页损失函数/Hinge损失函数</td>
</tr>
<tr><td>Hold-Out</td><td>留出法</td>
</tr>
<tr><td>Hyperparameter</td><td>超参数</td>
</tr>
<tr><td>Hyperparameter Optimization</td><td>超参数优化</td>
</tr>
<tr><td>Hypothesis</td><td>假设</td>
</tr>
<tr><td>Hypothesis Space</td><td>假设空间</td>
</tr>
<tr><td>Hypothesis Test</td><td>假设检验</td>
</tr>
<tr><td>Identity Matrix</td><td>单位矩阵</td>
</tr>
<tr><td>Imitation Learning</td><td>模仿学习</td>
</tr>
<tr><td>Importance Sampling</td><td>重要性采样</td>
</tr>
<tr><td>Improved Iterative Scaling</td><td>改进的迭代尺度法</td>
</tr>
<tr><td>Incremental Learning</td><td>增量学习</td>
</tr>
<tr><td>Independent and Identically Distributed</td><td>独立同分布</td>
</tr>
<tr><td>Indicator Function</td><td>指示函数</td>
</tr>
<tr><td>Individual Learner</td><td>个体学习器</td>
</tr>
<tr><td>Induction</td><td>归纳</td>
</tr>
<tr><td>Inductive Bias</td><td>归纳偏好</td>
</tr>
<tr><td>Inductive Learning</td><td>归纳学习</td>
</tr>
<tr><td>Inductive Logic Programming</td><td>归纳逻辑程序设计</td>
</tr>
<tr><td>Inference</td><td>推断</td>
</tr>
<tr><td>Information Entropy</td><td>信息熵</td>
</tr>
<tr><td>Information Gain</td><td>信息增益</td>
</tr>
<tr><td>Inner Product</td><td>内积</td>
</tr>
<tr><td>Instance</td><td>示例</td>
</tr>
<tr><td>Internal Covariate Shift</td><td>内部协变量偏移</td>
</tr>
<tr><td>Inverse Matrix</td><td>逆矩阵</td>
</tr>
<tr><td>Inverse Resolution</td><td>逆归结</td>
</tr>
<tr><td>Isometric Mapping</td><td>等度量映射</td>
</tr>
<tr><td>Jacobian Matrix</td><td>雅可比矩阵</td>
</tr>
<tr><td>Jensen Inequality</td><td>Jensen不等式</td>
</tr>
<tr><td>Joint Probability Distribution</td><td>联合概率分布</td>
</tr>
<tr><td>K-Armed Bandit Problem</td><td>k-摇臂老虎机</td>
</tr>
<tr><td>K-Fold Cross Validation</td><td>k 折交叉验证</td>
</tr>
<tr><td>Karush-Kuhn-Tucker Condition</td><td>KKT条件</td>
</tr>
<tr><td>Karush–Kuhn–Tucker</td><td>Karush–Kuhn–Tucker</td>
</tr>
<tr><td>Kernel Function</td><td>核函数</td>
</tr>
<tr><td>Kernel Method</td><td>核方法</td>
</tr>
<tr><td>Kernel Trick</td><td>核技巧</td>
</tr>
<tr><td>Kernelized Linear Discriminant Analysis</td><td>核线性判别分析</td>
</tr>
<tr><td>KL Divergence</td><td>KL散度</td>
</tr>
<tr><td>L-BFGS</td><td>L-BFGS</td>
</tr>
<tr><td>Label</td><td>标签</td>
</tr>
<tr><td>Label Space</td><td>标记空间</td>
</tr>
<tr><td>Lagrange Duality</td><td>拉格朗日对偶性</td>
</tr>
<tr><td>Lagrange Multiplier</td><td>拉格朗日乘子</td>
</tr>
<tr><td>Language Model</td><td>语言模型</td>
</tr>
<tr><td>Laplace Smoothing</td><td>拉普拉斯平滑</td>
</tr>
<tr><td>Laplacian Correction</td><td>拉普拉斯修正</td>
</tr>
<tr><td>Latent Dirichlet Allocation</td><td>潜在狄利克雷分配</td>
</tr>
<tr><td>Latent Semantic Analysis</td><td>潜在语义分析</td>
</tr>
<tr><td>Latent Variable</td><td>潜变量/隐变量</td>
</tr>
<tr><td>Law of Large Numbers</td><td>大数定律</td>
</tr>
<tr><td>Layer Normalization</td><td>层规范化</td>
</tr>
<tr><td>Lazy Learning</td><td>懒惰学习</td>
</tr>
<tr><td>Leaky Relu</td><td>泄漏修正线性单元/泄漏整流线性单元</td>
</tr>
<tr><td>Learner</td><td>学习器</td>
</tr>
<tr><td>Learning</td><td>学习</td>
</tr>
<tr><td>Learning By Analogy</td><td>类比学习</td>
</tr>
<tr><td>Learning Rate</td><td>学习率</td>
</tr>
<tr><td>Learning Vector Quantization</td><td>学习向量量化</td>
</tr>
<tr><td>Least Square Method</td><td>最小二乘法</td>
</tr>
<tr><td>Least Squares Regression Tree</td><td>最小二乘回归树</td>
</tr>
<tr><td>Left Singular Vector</td><td>左奇异向量</td>
</tr>
<tr><td>Likelihood</td><td>似然</td>
</tr>
<tr><td>Linear Chain Conditional Random Field</td><td>线性链条件随机场</td>
</tr>
<tr><td>Linear Classification Model</td><td>线性分类模型</td>
</tr>
<tr><td>Linear Classifier</td><td>线性分类器</td>
</tr>
<tr><td>Linear Dependence</td><td>线性相关</td>
</tr>
<tr><td>Linear Discriminant Analysis</td><td>线性判别分析</td>
</tr>
<tr><td>Linear Model</td><td>线性模型</td>
</tr>
<tr><td>Linear Regression</td><td>线性回归</td>
</tr>
<tr><td>Link Function</td><td>联系函数</td>
</tr>
<tr><td>Local Markov Property</td><td>局部马尔可夫性</td>
</tr>
<tr><td>Local Minima</td><td>局部极小</td>
</tr>
<tr><td>Local Minimum</td><td>局部极小</td>
</tr>
<tr><td>Local Representation</td><td>局部式表示/局部式表征</td>
</tr>
<tr><td>Log Likelihood</td><td>对数似然函数</td>
</tr>
<tr><td>Log Linear Model</td><td>对数线性模型</td>
</tr>
<tr><td>Log-Likelihood</td><td>对数似然</td>
</tr>
<tr><td>Log-Linear Regression</td><td>对数线性回归</td>
</tr>
<tr><td>Logistic Function</td><td>对数几率函数</td>
</tr>
<tr><td>Logistic Regression</td><td>对数几率回归</td>
</tr>
<tr><td>Logit</td><td>对数几率</td>
</tr>
<tr><td>Long Short Term Memory</td><td>长短期记忆</td>
</tr>
<tr><td>Long Short-Term Memory Network</td><td>长短期记忆网络</td>
</tr>
<tr><td>Loopy Belief Propagation</td><td>环状信念传播</td>
</tr>
<tr><td>Loss Function</td><td>损失函数</td>
</tr>
<tr><td>Low Rank Matrix Approximation</td><td>低秩矩阵近似</td>
</tr>
<tr><td>Machine Learning</td><td>机器学习</td>
</tr>
<tr><td>Macron-R</td><td>宏查全率</td>
</tr>
<tr><td>Manhattan Distance</td><td>曼哈顿距离</td>
</tr>
<tr><td>Manifold</td><td>流形</td>
</tr>
<tr><td>Manifold Assumption</td><td>流形假设</td>
</tr>
<tr><td>Manifold Learning</td><td>流形学习</td>
</tr>
<tr><td>Margin</td><td>间隔</td>
</tr>
<tr><td>Marginal Distribution</td><td>边缘分布</td>
</tr>
<tr><td>Marginal Independence</td><td>边缘独立性</td>
</tr>
<tr><td>Marginalization</td><td>边缘化</td>
</tr>
<tr><td>Markov Chain</td><td>马尔可夫链</td>
</tr>
<tr><td>Markov Chain Monte Carlo</td><td>马尔可夫链蒙特卡罗</td>
</tr>
<tr><td>Markov Decision Process</td><td>马尔可夫决策过程</td>
</tr>
<tr><td>Markov Network</td><td>马尔可夫网络</td>
</tr>
<tr><td>Markov Process</td><td>马尔可夫过程</td>
</tr>
<tr><td>Markov Random Field</td><td>马尔可夫随机场</td>
</tr>
<tr><td>Mask</td><td>掩码</td>
</tr>
<tr><td>Matrix</td><td>矩阵</td>
</tr>
<tr><td>Matrix Inversion</td><td>逆矩阵</td>
</tr>
<tr><td>Max Pooling</td><td>最大汇聚</td>
</tr>
<tr><td>Maximal Clique</td><td>最大团</td>
</tr>
<tr><td>Maximum Entropy Model</td><td>最大熵模型</td>
</tr>
<tr><td>Maximum Likelihood Estimation</td><td>极大似然估计</td>
</tr>
<tr><td>Maximum Margin</td><td>最大间隔</td>
</tr>
<tr><td>Mean Filed</td><td>平均场</td>
</tr>
<tr><td>Mean Pooling</td><td>平均汇聚</td>
</tr>
<tr><td>Mean Squared Error</td><td>均方误差</td>
</tr>
<tr><td>Mean-Field</td><td>平均场</td>
</tr>
<tr><td>Memory Network</td><td>记忆网络</td>
</tr>
<tr><td>Message Passing</td><td>消息传递</td>
</tr>
<tr><td>Metric Learning</td><td>度量学习</td>
</tr>
<tr><td>Micro-R</td><td>微查全率</td>
</tr>
<tr><td>Minibatch</td><td>小批量</td>
</tr>
<tr><td>Minimal Description Length</td><td>最小描述长度</td>
</tr>
<tr><td>Minimax Game</td><td>极小极大博弈</td>
</tr>
<tr><td>Minkowski Distance</td><td>闵可夫斯基距离</td>
</tr>
<tr><td>Mixture of Experts</td><td>混合专家模型</td>
</tr>
<tr><td>Mixture-of-Gaussian</td><td>高斯混合</td>
</tr>
<tr><td>Model</td><td>模型</td>
</tr>
<tr><td>Model Selection</td><td>模型选择</td>
</tr>
<tr><td>Momentum Method</td><td>动量法</td>
</tr>
<tr><td>Monte Carlo Method</td><td>蒙特卡罗方法</td>
</tr>
<tr><td>Moral Graph</td><td>端正图/道德图</td>
</tr>
<tr><td>Moralization</td><td>道德化</td>
</tr>
<tr><td>Multi-Class Classification</td><td>多分类</td>
</tr>
<tr><td>Multi-Head Attention</td><td>多头注意力</td>
</tr>
<tr><td>Multi-Head Self-Attention</td><td>多头自注意力</td>
</tr>
<tr><td>Multi-Kernel Learning</td><td>多核学习</td>
</tr>
<tr><td>Multi-Label Learning</td><td>多标记学习</td>
</tr>
<tr><td>Multi-Layer Feedforward Neural Networks</td><td>多层前馈神经网络</td>
</tr>
<tr><td>Multi-Layer Perceptron</td><td>多层感知机</td>
</tr>
<tr><td>Multinomial Distribution</td><td>多项分布</td>
</tr>
<tr><td>Multiple Dimensional Scaling</td><td>多维缩放</td>
</tr>
<tr><td>Multiple Linear Regression</td><td>多元线性回归</td>
</tr>
<tr><td>Multitask Learning</td><td>多任务学习</td>
</tr>
<tr><td>Multivariate Normal Distribution</td><td>多元正态分布</td>
</tr>
<tr><td>Mutual Information</td><td>互信息</td>
</tr>
<tr><td>N-Gram Model</td><td>N元模型</td>
</tr>
<tr><td>Naive Bayes Classifier</td><td>朴素贝叶斯分类器</td>
</tr>
<tr><td>Naive Bayes</td><td>朴素贝叶斯</td>
</tr>
<tr><td>Nearest Neighbor Classifier</td><td>最近邻分类器</td>
</tr>
<tr><td>Negative Log Likelihood</td><td>负对数似然函数</td>
</tr>
<tr><td>Neighbourhood Component Analysis</td><td>近邻成分分析</td>
</tr>
<tr><td>Net Input</td><td>净输入</td>
</tr>
<tr><td>Neural Network</td><td>神经网络</td>
</tr>
<tr><td>Neural Turing Machine</td><td>神经图灵机</td>
</tr>
<tr><td>Neuron</td><td>神经元</td>
</tr>
<tr><td>Newton Method</td><td>牛顿法</td>
</tr>
<tr><td>No Free Lunch Theorem</td><td>没有免费午餐定理</td>
</tr>
<tr><td>Noise-Contrastive Estimation</td><td>噪声对比估计</td>
</tr>
<tr><td>Nominal Attribute</td><td>列名属性</td>
</tr>
<tr><td>Non-Convex Optimization</td><td>非凸优化</td>
</tr>
<tr><td>Non-Metric Distance</td><td>非度量距离</td>
</tr>
<tr><td>Non-Negative Matrix Factorization</td><td>非负矩阵分解</td>
</tr>
<tr><td>Non-Ordinal Attribute</td><td>无序属性</td>
</tr>
<tr><td>Norm</td><td>范数</td>
</tr>
<tr><td>Normal Distribution</td><td>正态分布</td>
</tr>
<tr><td>Normalization</td><td>规范化</td>
</tr>
<tr><td>Nuclear Norm</td><td>核范数</td>
</tr>
<tr><td>Number of Epochs</td><td>轮数</td>
</tr>
<tr><td>Numerical Attribute</td><td>数值属性</td>
</tr>
<tr><td>Object Detection</td><td>目标检测</td>
</tr>
<tr><td>Oblique Decision Tree</td><td>斜决策树</td>
</tr>
<tr><td>Occam's Razor</td><td>奥卡姆剃刀</td>
</tr>
<tr><td>Odds</td><td>几率</td>
</tr>
<tr><td>Off-Policy</td><td>异策略</td>
</tr>
<tr><td>On-Policy</td><td>同策略</td>
</tr>
<tr><td>One-Dependent Estimator</td><td>独依赖估计</td>
</tr>
<tr><td>One-Hot</td><td>独热</td>
</tr>
<tr><td>Online Learning</td><td>在线学习</td>
</tr>
<tr><td>Optimizer</td><td>优化器</td>
</tr>
<tr><td>Ordinal Attribute</td><td>有序属性</td>
</tr>
<tr><td>Orthogonal</td><td>正交</td>
</tr>
<tr><td>Orthogonal Matrix</td><td>正交矩阵</td>
</tr>
<tr><td>Out-Of-Bag Estimate</td><td>包外估计</td>
</tr>
<tr><td>Outlier</td><td>异常点</td>
</tr>
<tr><td>Over-Parameterized</td><td>过度参数化</td>
</tr>
<tr><td>Overfitting</td><td>过拟合</td>
</tr>
<tr><td>Oversampling</td><td>过采样</td>
</tr>
<tr><td>Pac-Learnable</td><td>PAC可学习</td>
</tr>
<tr><td>Padding</td><td>填充</td>
</tr>
<tr><td>Pairwise Markov Property</td><td>成对马尔可夫性</td>
</tr>
<tr><td>Parallel Distributed Processing</td><td>分布式并行处理</td>
</tr>
<tr><td>Parameter</td><td>参数</td>
</tr>
<tr><td>Parameter Estimation</td><td>参数估计</td>
</tr>
<tr><td>Parameter Space</td><td>参数空间</td>
</tr>
<tr><td>Parameter Tuning</td><td>调参</td>
</tr>
<tr><td>Parametric ReLU</td><td>参数化修正线性单元/参数化整流线性单元</td>
</tr>
<tr><td>Part-Of-Speech Tagging</td><td>词性标注</td>
</tr>
<tr><td>Partial Derivative</td><td>偏导数</td>
</tr>
<tr><td>Partially Observable Markov Decision Processes</td><td>部分可观测马尔可夫决策过程</td>
</tr>
<tr><td>Partition Function</td><td>配分函数</td>
</tr>
<tr><td>Perceptron</td><td>感知机</td>
</tr>
<tr><td>Performance Measure</td><td>性能度量</td>
</tr>
<tr><td>Perplexity</td><td>困惑度</td>
</tr>
<tr><td>Pointer Network</td><td>指针网络</td>
</tr>
<tr><td>Policy</td><td>策略</td>
</tr>
<tr><td>Policy Gradient</td><td>策略梯度</td>
</tr>
<tr><td>Policy Iteration</td><td>策略迭代</td>
</tr>
<tr><td>Polynomial Kernel Function</td><td>多项式核函数</td>
</tr>
<tr><td>Pooling</td><td>汇聚</td>
</tr>
<tr><td>Pooling Layer</td><td>汇聚层</td>
</tr>
<tr><td>Positive Definite Matrix</td><td>正定矩阵</td>
</tr>
<tr><td>Post-Pruning</td><td>后剪枝</td>
</tr>
<tr><td>Potential Function</td><td>势函数</td>
</tr>
<tr><td>Power Method</td><td>幂法</td>
</tr>
<tr><td>Pre-Training</td><td>预训练</td>
</tr>
<tr><td>Precision</td><td>查准率/准确率</td>
</tr>
<tr><td>Prepruning</td><td>预剪枝</td>
</tr>
<tr><td>Primal Problem</td><td>主问题</td>
</tr>
<tr><td>Primary Visual Cortex</td><td>初级视觉皮层</td>
</tr>
<tr><td>Principal Component Analysis</td><td>主成分分析</td>
</tr>
<tr><td>Prior</td><td>先验</td>
</tr>
<tr><td>Probabilistic Context-Free Grammar</td><td>概率上下文无关文法</td>
</tr>
<tr><td>Probabilistic Graphical Model</td><td>概率图模型</td>
</tr>
<tr><td>Probabilistic Model</td><td>概率模型</td>
</tr>
<tr><td>Probability Density Function</td><td>概率密度函数</td>
</tr>
<tr><td>Probability Distribution</td><td>概率分布</td>
</tr>
<tr><td>Probably Approximately Correct</td><td>概率近似正确</td>
</tr>
<tr><td>Proposal Distribution</td><td>提议分布</td>
</tr>
<tr><td>Prototype-Based Clustering</td><td>原型聚类</td>
</tr>
<tr><td>Proximal Gradient Descent</td><td>近端梯度下降</td>
</tr>
<tr><td>Pruning</td><td>剪枝</td>
</tr>
<tr><td>Quadratic Loss Function</td><td>平方损失函数</td>
</tr>
<tr><td>Quadratic Programming</td><td>二次规划</td>
</tr>
<tr><td>Quasi Newton Method</td><td>拟牛顿法</td>
</tr>
<tr><td>Radial Basis Function</td><td>径向基函数</td>
</tr>
<tr><td>Random Forest</td><td>随机森林</td>
</tr>
<tr><td>Random Sampling</td><td>随机采样</td>
</tr>
<tr><td>Random Search</td><td>随机搜索</td>
</tr>
<tr><td>Random Variable</td><td>随机变量</td>
</tr>
<tr><td>Random Walk</td><td>随机游走</td>
</tr>
<tr><td>Recall</td><td>查全率/召回率</td>
</tr>
<tr><td>Receptive Field</td><td>感受野</td>
</tr>
<tr><td>Reconstruction Error</td><td>重构误差</td>
</tr>
<tr><td>Rectified Linear Unit</td><td>修正线性单元/整流线性单元</td>
</tr>
<tr><td>Recurrent Neural Network</td><td>循环神经网络</td>
</tr>
<tr><td>Recursive Neural Network</td><td>递归神经网络</td>
</tr>
<tr><td>Regression</td><td>回归</td>
</tr>
<tr><td>Regularization</td><td>正则化</td>
</tr>
<tr><td>Regularizer</td><td>正则化项</td>
</tr>
<tr><td>Reinforcement Learning</td><td>强化学习</td>
</tr>
<tr><td>Relative Entropy</td><td>相对熵</td>
</tr>
<tr><td>Reparameterization</td><td>再参数化/重参数化</td>
</tr>
<tr><td>Representation</td><td>表示</td>
</tr>
<tr><td>Representation Learning</td><td>表示学习</td>
</tr>
<tr><td>Representer Theorem</td><td>表示定理</td>
</tr>
<tr><td>Reproducing Kernel Hilbert Space</td><td>再生核希尔伯特空间</td>
</tr>
<tr><td>Rescaling</td><td>再缩放</td>
</tr>
<tr><td>Reset Gate</td><td>重置门</td>
</tr>
<tr><td>Residual Connection</td><td>残差连接</td>
</tr>
<tr><td>Residual Network</td><td>残差网络</td>
</tr>
<tr><td>Restricted Boltzmann Machine</td><td>受限玻尔兹曼机</td>
</tr>
<tr><td>Reward</td><td>奖励</td>
</tr>
<tr><td>Ridge Regression</td><td>岭回归</td>
</tr>
<tr><td>Right Singular Vector</td><td>右奇异向量</td>
</tr>
<tr><td>Risk</td><td>风险</td>
</tr>
<tr><td>Robustness</td><td>稳健性</td>
</tr>
<tr><td>Root Node</td><td>根结点</td>
</tr>
<tr><td>Rule Learning</td><td>规则学习</td>
</tr>
<tr><td>Saddle Point</td><td>鞍点</td>
</tr>
<tr><td>Sample</td><td>样本</td>
</tr>
<tr><td>Sample Complexity</td><td>样本复杂度</td>
</tr>
<tr><td>Sample Space</td><td>样本空间</td>
</tr>
<tr><td>Scalar</td><td>标量</td>
</tr>
<tr><td>Selective Ensemble</td><td>选择性集成</td>
</tr>
<tr><td>Self Information</td><td>自信息</td>
</tr>
<tr><td>Self-Attention</td><td>自注意力</td>
</tr>
<tr><td>Self-Organizing Map</td><td>自组织映射网</td>
</tr>
<tr><td>Self-Training</td><td>自训练</td>
</tr>
<tr><td>Semi-Definite Programming</td><td>半正定规划</td>
</tr>
<tr><td>Semi-Naive Bayes Classifiers</td><td>半朴素贝叶斯分类器</td>
</tr>
<tr><td>Semi-Restricted Boltzmann Machine</td><td>半受限玻尔兹曼机</td>
</tr>
<tr><td>Semi-Supervised Clustering</td><td>半监督聚类</td>
</tr>
<tr><td>Semi-Supervised Learning</td><td>半监督学习</td>
</tr>
<tr><td>Semi-Supervised Support Vector Machine</td><td>半监督支持向量机</td>
</tr>
<tr><td>Sentiment Analysis</td><td>情感分析</td>
</tr>
<tr><td>Separating Hyperplane</td><td>分离超平面</td>
</tr>
<tr><td>Sequential Covering</td><td>序贯覆盖</td>
</tr>
<tr><td>Sigmoid Belief Network</td><td>Sigmoid信念网络</td>
</tr>
<tr><td>Sigmoid Function</td><td>Sigmoid函数</td>
</tr>
<tr><td>Signed Distance</td><td>带符号距离</td>
</tr>
<tr><td>Similarity Measure</td><td>相似度度量</td>
</tr>
<tr><td>Simulated Annealing</td><td>模拟退火</td>
</tr>
<tr><td>Simultaneous Localization And Mapping</td><td>即时定位与地图构建</td>
</tr>
<tr><td>Singular Value</td><td>奇异值</td>
</tr>
<tr><td>Singular Value Decomposition</td><td>奇异值分解</td>
</tr>
<tr><td>Skip-Gram Model</td><td>跳元模型</td>
</tr>
<tr><td>Smoothing</td><td>平滑</td>
</tr>
<tr><td>Soft Margin</td><td>软间隔</td>
</tr>
<tr><td>Soft Margin Maximization</td><td>软间隔最大化</td>
</tr>
<tr><td>Softmax</td><td>Softmax/软最大化</td>
</tr>
<tr><td>Softmax Function</td><td>Softmax函数/软最大化函数</td>
</tr>
<tr><td>Softmax Regression</td><td>Softmax回归/软最大化回归</td>
</tr>
<tr><td>Softplus Function</td><td>Softplus函数</td>
</tr>
<tr><td>Span</td><td>张成子空间</td>
</tr>
<tr><td>Sparse Coding</td><td>稀疏编码</td>
</tr>
<tr><td>Sparse Representation</td><td>稀疏表示</td>
</tr>
<tr><td>Sparsity</td><td>稀疏性</td>
</tr>
<tr><td>Specialization</td><td>特化</td>
</tr>
<tr><td>Splitting Variable</td><td>切分变量</td>
</tr>
<tr><td>Squashing Function</td><td>挤压函数</td>
</tr>
<tr><td>Standard Normal Distribution</td><td>标准正态分布</td>
</tr>
<tr><td>State</td><td>状态</td>
</tr>
<tr><td>State Value Function</td><td>状态值函数</td>
</tr>
<tr><td>State-Action Value Function</td><td>状态-动作值函数</td>
</tr>
<tr><td>Stationary Distribution</td><td>平稳分布</td>
</tr>
<tr><td>Stationary Point</td><td>驻点</td>
</tr>
<tr><td>Statistical Learning</td><td>统计学习</td>
</tr>
<tr><td>Steepest Descent</td><td>最速下降法</td>
</tr>
<tr><td>Stochastic Gradient Descent</td><td>随机梯度下降</td>
</tr>
<tr><td>Stochastic Matrix</td><td>随机矩阵</td>
</tr>
<tr><td>Stochastic Process</td><td>随机过程</td>
</tr>
<tr><td>Stratified Sampling</td><td>分层采样</td>
</tr>
<tr><td>Stride</td><td>步幅</td>
</tr>
<tr><td>Structural Risk</td><td>结构风险</td>
</tr>
<tr><td>Structural Risk Minimization</td><td>结构风险最小化</td>
</tr>
<tr><td>Subsample</td><td>子采样</td>
</tr>
<tr><td>Subsampling</td><td>下采样</td>
</tr>
<tr><td>Subset Search</td><td>子集搜索</td>
</tr>
<tr><td>Subspace</td><td>子空间</td>
</tr>
<tr><td>Supervised Learning</td><td>监督学习</td>
</tr>
<tr><td>Support Vector</td><td>支持向量</td>
</tr>
<tr><td>Support Vector Expansion</td><td>支持向量展式</td>
</tr>
<tr><td>Support Vector Machine</td><td>支持向量机</td>
</tr>
<tr><td>Surrogat Loss</td><td>替代损失</td>
</tr>
<tr><td>Surrogate Function</td><td>替代函数</td>
</tr>
<tr><td>Surrogate Loss Function</td><td>代理损失函数</td>
</tr>
<tr><td>Symbolism</td><td>符号主义</td>
</tr>
<tr><td>Tangent Propagation</td><td>正切传播</td>
</tr>
<tr><td>Teacher Forcing</td><td>强制教学</td>
</tr>
<tr><td>Temporal-Difference Learning</td><td>时序差分学习</td>
</tr>
<tr><td>Tensor</td><td>张量</td>
</tr>
<tr><td>Test Error</td><td>测试误差</td>
</tr>
<tr><td>Test Sample</td><td>测试样本</td>
</tr>
<tr><td>Test Set</td><td>测试集</td>
</tr>
<tr><td>Threshold</td><td>阈值</td>
</tr>
<tr><td>Threshold Logic Unit</td><td>阈值逻辑单元</td>
</tr>
<tr><td>Threshold-Moving</td><td>阈值移动</td>
</tr>
<tr><td>Tied Weight</td><td>捆绑权重</td>
</tr>
<tr><td>Tikhonov Regularization</td><td>Tikhonov正则化</td>
</tr>
<tr><td>Time Delay Neural Network</td><td>时延神经网络</td>
</tr>
<tr><td>Time Homogenous Markov Chain</td><td>时间齐次马尔可夫链</td>
</tr>
<tr><td>Time Step</td><td>时间步</td>
</tr>
<tr><td>Token</td><td>词元</td>
</tr>
<tr><td>Token</td><td>词元</td>
</tr>
<tr><td>Tokenization</td><td>词元化</td>
</tr>
<tr><td>Tokenizer</td><td>词元分析器</td>
</tr>
<tr><td>Topic Model</td><td>话题模型</td>
</tr>
<tr><td>Topic Modeling</td><td>话题分析</td>
</tr>
<tr><td>Trace</td><td>迹</td>
</tr>
<tr><td>Training</td><td>训练</td>
</tr>
<tr><td>Training Error</td><td>训练误差</td>
</tr>
<tr><td>Training Sample</td><td>训练样本</td>
</tr>
<tr><td>Training Set</td><td>训练集</td>
</tr>
<tr><td>Transductive Learning</td><td>直推学习</td>
</tr>
<tr><td>Transductive Transfer Learning</td><td>直推迁移学习</td>
</tr>
<tr><td>Transfer Learning</td><td>迁移学习</td>
</tr>
<tr><td>Transformer</td><td>Transformer</td>
</tr>
<tr><td>Transformer Model</td><td>Transformer模型</td>
</tr>
<tr><td>Transpose</td><td>转置</td>
</tr>
<tr><td>Transposed Convolution</td><td>转置卷积</td>
</tr>
<tr><td>Trial And Error</td><td>试错</td>
</tr>
<tr><td>Trigram</td><td>三元语法</td>
</tr>
<tr><td>Turing Machine</td><td>图灵机</td>
</tr>
<tr><td>Underfitting</td><td>欠拟合</td>
</tr>
<tr><td>Undersampling</td><td>欠采样</td>
</tr>
<tr><td>Undirected Graphical Model</td><td>无向图模型</td>
</tr>
<tr><td>Uniform Distribution</td><td>均匀分布</td>
</tr>
<tr><td>Unigram</td><td>一元语法</td>
</tr>
<tr><td>Unit</td><td>单元</td>
</tr>
<tr><td>Universal Approximation Theorem</td><td>通用近似定理</td>
</tr>
<tr><td>Universal Approximator</td><td>通用近似器</td>
</tr>
<tr><td>Universal Function Approximator</td><td>通用函数近似器</td>
</tr>
<tr><td>Unknown Token</td><td>未知词元</td>
</tr>
<tr><td>Unsupervised Layer-Wise Training</td><td>无监督逐层训练</td>
</tr>
<tr><td>Unsupervised Learning</td><td>无监督学习</td>
</tr>
<tr><td>Update Gate</td><td>更新门</td>
</tr>
<tr><td>Upsampling</td><td>上采样</td>
</tr>
<tr><td>V-Structure</td><td>V型结构</td>
</tr>
<tr><td>Validation Set</td><td>验证集</td>
</tr>
<tr><td>Validity Index</td><td>有效性指标</td>
</tr>
<tr><td>Value Function Approximation</td><td>值函数近似</td>
</tr>
<tr><td>Value Iteration</td><td>值迭代</td>
</tr>
<tr><td>Vanishing Gradient Problem</td><td>梯度消失问题</td>
</tr>
<tr><td>Vapnik-Chervonenkis Dimension</td><td>VC维</td>
</tr>
<tr><td>Variable Elimination</td><td>变量消去</td>
</tr>
<tr><td>Variance</td><td>方差</td>
</tr>
<tr><td>Variational Autoencoder</td><td>变分自编码器</td>
</tr>
<tr><td>Variational Inference</td><td>变分推断</td>
</tr>
<tr><td>Vector</td><td>向量</td>
</tr>
<tr><td>Vector Space Model</td><td>向量空间模型</td>
</tr>
<tr><td>Version Space</td><td>版本空间</td>
</tr>
<tr><td>Viterbi Algorithm</td><td>维特比算法</td>
</tr>
<tr><td>Vocabulary</td><td>词表</td>
</tr>
<tr><td>Warp</td><td>线程束</td>
</tr>
<tr><td>Weak Learner</td><td>弱学习器</td>
</tr>
<tr><td>Weakly Supervised Learning</td><td>弱监督学习</td>
</tr>
<tr><td>Weight</td><td>权重</td>
</tr>
<tr><td>Weight Decay</td><td>权重衰减</td>
</tr>
<tr><td>Weight Sharing</td><td>权共享</td>
</tr>
<tr><td>Weighted Voting</td><td>加权投票</td>
</tr>
<tr><td>Whitening</td><td>白化</td>
</tr>
<tr><td>Winner-Take-All</td><td>胜者通吃</td>
</tr>
<tr><td>Within-Class Scatter Matrix</td><td>类内散度矩阵</td>
</tr>
<tr><td>Word Embedding</td><td>词嵌入</td>
</tr>
<tr><td>Word Sense Disambiguation</td><td>词义消歧</td>
</tr>
<tr><td>Word Vector</td><td>词向量</td>
</tr>
<tr><td>Zero Padding</td><td>零填充</td>
</tr>
<tr><td>Zero-Shot Learning</td><td>零试学习</td>
</tr>
<tr><td>Zipf's Law</td><td>齐普夫定律</td>
</tr>
</tbody></table><br>
<a href="https://mp.weixin.qq.com/s/8oDBsZ26BjEfwaHVOlmk0Q"><font color="#9a9a9a">来源：<u>Python数据科学</u></font></a><br>
