https://terms-en.ai-term-hub.com/zh/tags/statistics/Recent content in Statistics on 中文AI术语词典Hugozh-cnSat, 18 Jul 2026 11:44:45 +0000https://terms-en.ai-term-hub.com/zh/terms/spike_and_slab_regression/Sat, 18 Jul 2026 11:35:05 +0000https://terms-en.ai-term-hub.com/zh/terms/spike_and_slab_regression/<h2 id="definition">Definition</h2> <p>尖峰-厚尾回归（Spike-and-slab regression）是一种用于变量选择和稀疏建模的贝叶斯统计技术。它采用由两个部分组成的混合先验分布：一个‘尖峰’（spike，通常集中在零点附近以表示系数为零）和一个‘厚尾’（slab，较宽的分布以表示非零系数）。这种方法通过概率方式自动进行特征选择，特别适用于高维数据中识别重要变量的场景。</p> <h3 id="summary">Summary</h3> <p>一种贝叶斯变量选择方法，使用混合先验分布来区分系数为零和非零的情况。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>贝叶斯推断</li> <li>稀疏建模</li> <li>混合先验</li> <li>变量选择</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>高维基因组数据分析</li> <li>金融风险因子识别</li> <li>预测模型中的特征选择</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/lasso-l1%E6%AD%A3%E5%88%99%E5%8C%96%E5%9B%9E%E5%BD%92/">Lasso (L1正则化回归)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/ridge-regression-%E5%B2%AD%E5%9B%9E%E5%BD%92-l2%E6%AD%A3%E5%88%99%E5%8C%96/">Ridge Regression (岭回归/L2正则化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bayesian-linear-regression-%E8%B4%9D%E5%8F%B6%E6%96%AF%E7%BA%BF%E6%80%A7%E5%9B%9E%E5%BD%92/">Bayesian Linear Regression (贝叶斯线性回归)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/sparsity-%E7%A8%80%E7%96%8F%E6%80%A7/">Sparsity (稀疏性)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/regularization/Sat, 18 Jul 2026 11:32:02 +0000https://terms-en.ai-term-hub.com/zh/terms/regularization/<h2 id="definition">Definition</h2> <p>正则化是机器学习中一个关键概念，旨在不显著增加训练误差的情况下减少泛化误差。它通过阻止模型学习过于复杂的模式来发挥作用。</p> <h3 id="summary">Summary</h3> <p>一组在训练过程中使用的技术，通过对损失函数添加惩罚项或限制模型复杂度来防止过拟合。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>过拟合</li> <li>偏差-方差权衡</li> <li>L1/L2 惩罚</li> <li>丢弃法 (Dropout)</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>深度神经网络训练</li> <li>线性回归模型</li> <li>防止小数据集上的记忆效应</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> sklearn.linear_model <span style="color:#f92672">import</span> Ridge </span></span><span style="display:flex;"><span>model <span style="color:#f92672">=</span> Ridge(alpha<span style="color:#f92672">=</span><span style="color:#ae81ff">1.0</span>) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/overfitting-%E8%BF%87%E6%8B%9F%E5%90%88/">Overfitting (过拟合)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/underfitting-%E6%AC%A0%E6%8B%9F%E5%90%88/">Underfitting (欠拟合)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/cross-validation-%E4%BA%A4%E5%8F%89%E9%AA%8C%E8%AF%81/">Cross-validation (交叉验证)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/hyperparameter-tuning-%E8%B6%85%E5%8F%82%E6%95%B0%E8%B0%83%E4%BC%98/">Hyperparameter tuning (超参数调优)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/rademacher_complexity/Sat, 18 Jul 2026 11:31:50 +0000https://terms-en.ai-term-hub.com/zh/terms/rademacher_complexity/<h2 id="definition">Definition</h2> <p>Rademacher 复杂度评估假设类与随机标签（噪声）的相关程度。它作为模型容量或灵活性的代理指标。较低的复杂度通常意味着更好的泛化能力。</p> <h3 id="summary">Summary</h3> <p>学习理论中用于量化函数类丰富程度的统计度量，通过其拟合随机噪声的能力来评估。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>泛化界</li> <li>模型容量</li> <li>过拟合</li> <li>统计学习理论</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>支持向量机 (SVM) 的理论分析</li> <li>推导收敛速率</li> <li>比较不同模型类</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/vc-%E7%BB%B4-vapnik-chervonenkis-dimension/">VC 维 (Vapnik-Chervonenkis dimension)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%B3%9B%E5%8C%96%E8%AF%AF%E5%B7%AE-generalization-error/">泛化误差 (Generalization error)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/pac-%E5%AD%A6%E4%B9%A0-probably-approximately-correct-learning/">PAC 学习 (Probably Approximately Correct learning)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%81%8F%E5%B7%AE-%E6%96%B9%E5%B7%AE%E6%9D%83%E8%A1%A1-bias-variance-tradeoff/">偏差-方差权衡 (Bias-variance tradeoff)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/phi_coefficient/Sat, 18 Jul 2026 11:30:05 +0000https://terms-en.ai-term-hub.com/zh/terms/phi_coefficient/<h2 id="definition">Definition</h2> <p>Phi系数（φ）是用于衡量两个二元变量之间关联程度的指标，可视为二值变量的皮尔逊相关系数。其取值范围为-1到+1，其中0表示无关联。</p> <h3 id="summary">Summary</h3> <p>衡量两个二元变量之间关联程度的统计指标。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>二元变量</li> <li>相关性</li> <li>列联表</li> <li>关联强度</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>评估二元分类器在准确率之外的性能</li> <li>分析具有“是/否”回复的调查数据中的关系</li> <li>在包含分类输入的数据集中进行特征选择</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">import</span> numpy <span style="color:#66d9ef">as</span> np </span></span><span style="display:flex;"><span><span style="color:#f92672">from</span> scipy.stats <span style="color:#f92672">import</span> chi2_contingency </span></span><span style="display:flex;"><span><span style="color:#75715e"># Example: Calculate phi coefficient from a 2x2 confusion matrix</span> </span></span><span style="display:flex;"><span>tn, fp, fn, tp <span style="color:#f92672">=</span> <span style="color:#ae81ff">90</span>, <span style="color:#ae81ff">10</span>, <span style="color:#ae81ff">5</span>, <span style="color:#ae81ff">95</span> </span></span><span style="display:flex;"><span>matrix <span style="color:#f92672">=</span> [[tn, fp], [fn, tp]] </span></span><span style="display:flex;"><span>chi2, p, dof, expected <span style="color:#f92672">=</span> chi2_contingency(matrix) </span></span><span style="display:flex;"><span>phi <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>sqrt(chi2 <span style="color:#f92672">/</span> (tn <span style="color:#f92672">+</span> fp <span style="color:#f92672">+</span> fn <span style="color:#f92672">+</span> tp)) </span></span><span style="display:flex;"><span>print(<span style="color:#e6db74">f</span><span style="color:#e6db74">&#39;Phi coefficient: </span><span style="color:#e6db74">{</span>phi<span style="color:#e6db74">:</span><span style="color:#e6db74">.3f</span><span style="color:#e6db74">}</span><span style="color:#e6db74">&#39;</span>) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/cramer-s-v-%E5%85%8B%E6%8B%89%E9%BB%98v%E7%B3%BB%E6%95%B0/">Cramer&rsquo;s V (克拉默V系数)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/pearson-correlation-%E7%9A%AE%E5%B0%94%E9%80%8A%E7%9B%B8%E5%85%B3%E7%B3%BB%E6%95%B0/">Pearson correlation (皮尔逊相关系数)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/confusion-matrix-%E6%B7%B7%E6%B7%86%E7%9F%A9%E9%98%B5/">Confusion matrix (混淆矩阵)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mutual-information-%E4%BA%92%E4%BF%A1%E6%81%AF/">Mutual information (互信息)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/perception_error_model/Sat, 18 Jul 2026 11:29:41 +0000https://terms-en.ai-term-hub.com/zh/terms/perception_error_model/<h2 id="definition">Definition</h2> <p>感知误差模型描述了观察到的感官数据与真实情况之间的差异，考虑了噪声、遮挡或传感器限制等因素。通过对这些误差进行建模，人工智能系统可以</p> <h3 id="summary">Summary</h3> <p>用于量化和校正感官数据解释中不准确性的统计或算法框架。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>噪声建模</li> <li>不确定性量化</li> <li>误差校正</li> <li>贝叶斯推理</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>激光雷达数据平滑</li> <li>相机校准调整</li> <li>容错机器人技术</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/kalman-filter-%E5%8D%A1%E5%B0%94%E6%9B%BC%E6%BB%A4%E6%B3%A2/">Kalman filter (卡尔曼滤波)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/sensor-noise-%E4%BC%A0%E6%84%9F%E5%99%A8%E5%99%AA%E5%A3%B0/">Sensor noise (传感器噪声)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/robust-estimation-%E9%B2%81%E6%A3%92%E4%BC%B0%E8%AE%A1/">Robust estimation (鲁棒估计)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/linear_predictor_function/Sat, 18 Jul 2026 11:24:20 +0000https://terms-en.ai-term-hub.com/zh/terms/linear_predictor_function/<h2 id="definition">Definition</h2> <p>在统计建模和机器学习中，线性预测函数表示输入特征的加权和加上偏置项。它是广义线性模型（GLM）的核心组成部分。</p> <h3 id="summary">Summary</h3> <p>一种计算输入变量的线性组合以预测结果的数学函数。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>加权和</li> <li>偏置项</li> <li>广义线性模型</li> <li>特征系数</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>线性回归分析</li> <li>逻辑回归分类</li> <li>支持向量机（核技巧上下文）</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">import</span> numpy <span style="color:#66d9ef">as</span> np </span></span><span style="display:flex;"><span>X <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>array([[<span style="color:#ae81ff">1</span>, <span style="color:#ae81ff">2</span>], [<span style="color:#ae81ff">3</span>, <span style="color:#ae81ff">4</span>]]) </span></span><span style="display:flex;"><span>w <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>array([<span style="color:#ae81ff">0.5</span>, <span style="color:#ae81ff">1.0</span>]) </span></span><span style="display:flex;"><span>b <span style="color:#f92672">=</span> <span style="color:#ae81ff">0.1</span> </span></span><span style="display:flex;"><span>prediction <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>dot(X, w) <span style="color:#f92672">+</span> b </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/regression_coefficients-%E5%9B%9E%E5%BD%92%E7%B3%BB%E6%95%B0/">regression_coefficients (回归系数)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bias_intercept-%E5%81%8F%E7%BD%AE-%E6%88%AA%E8%B7%9D/">bias_intercept (偏置/截距)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/feature_engineering-%E7%89%B9%E5%BE%81%E5%B7%A5%E7%A8%8B/">feature_engineering (特征工程)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/generalized_linear_model-%E5%B9%BF%E4%B9%89%E7%BA%BF%E6%80%A7%E6%A8%A1%E5%9E%8B/">generalized_linear_model (广义线性模型)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/leave_one_out_cross_validation/Sat, 18 Jul 2026 11:24:07 +0000https://terms-en.ai-term-hub.com/zh/terms/leave_one_out_cross_validation/<h2 id="definition">Definition</h2> <p>留一法交叉验证（LOOCV）是k折交叉验证的一种特殊情况，其中k等于数据集中的样本数量。它提供了对模型性能的近乎无偏的估计，因为每次训练都使用了尽可能多的数据，从而最大限度地减少了偏差。然而，这种方法计算成本较高，因为需要为每个样本重新训练模型。</p> <h3 id="summary">Summary</h3> <p>一种严格的重采样技术，模型在除一个样本外的所有样本上进行训练，并在该单个保留样本上进行测试，对每个数据点重复此过程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>重采样</li> <li>模型评估</li> <li>偏差-方差权衡</li> <li>计算成本</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>在小规模医疗数据集上评估模型</li> <li>数据稀缺时的超参数调优</li> <li>严格比较算法性能</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> sklearn.model_selection <span style="color:#f92672">import</span> LeaveOneOut </span></span><span style="display:flex;"><span>loo <span style="color:#f92672">=</span> LeaveOneOut() </span></span><span style="display:flex;"><span><span style="color:#66d9ef">for</span> train_index, test_index <span style="color:#f92672">in</span> loo<span style="color:#f92672">.</span>split(X): </span></span><span style="display:flex;"><span> X_train, X_test <span style="color:#f92672">=</span> X[train_index], X[test_index] </span></span><span style="display:flex;"><span> y_train, y_test <span style="color:#f92672">=</span> y[train_index], y[test_index] </span></span><span style="display:flex;"><span> model<span style="color:#f92672">.</span>fit(X_train, y_train) </span></span><span style="display:flex;"><span> score <span style="color:#f92672">=</span> model<span style="color:#f92672">.</span>score(X_test, y_test) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/k-fold-cross-validation-k%E6%8A%98%E4%BA%A4%E5%8F%89%E9%AA%8C%E8%AF%81/">k-fold cross-validation (k折交叉验证)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/train_test_split-%E8%AE%AD%E7%BB%83%E9%9B%86-%E6%B5%8B%E8%AF%95%E9%9B%86%E5%88%92%E5%88%86/">train_test_split (训练集-测试集划分)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bootstrap-%E8%87%AA%E5%8A%A9%E6%B3%95/">bootstrap (自助法)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/cross_validation_score-%E4%BA%A4%E5%8F%89%E9%AA%8C%E8%AF%81%E5%BE%97%E5%88%86/">cross_validation_score (交叉验证得分)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/life_time_of_correlation/Sat, 18 Jul 2026 11:24:07 +0000https://terms-en.ai-term-hub.com/zh/terms/life_time_of_correlation/<h2 id="definition">Definition</h2> <p>在动态系统和时间序列分析中，相关性寿命衡量两个变量保持显著统计依赖性的持续时间。这一概念对于理解模型随时间推移的性能下降至关重要，特别是在金融、气象和物联网等领域，其中变量之间的关系可能随时间迅速变化。</p> <h3 id="summary">Summary</h3> <p>一种指标，用于估算变量之间的统计关系在因概念漂移或环境变化而衰减之前保持稳定多长时间。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>概念漂移</li> <li>时间序列分析</li> <li>模型衰减</li> <li>统计稳定性</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>金融市场预测建模</li> <li>传感器网络维护计划</li> <li>自适应控制系统</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/concept_drift-%E6%A6%82%E5%BF%B5%E6%BC%82%E7%A7%BB/">concept_drift (概念漂移)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/stationarity-%E5%B9%B3%E7%A8%B3%E6%80%A7/">stationarity (平稳性)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/time_series-%E6%97%B6%E9%97%B4%E5%BA%8F%E5%88%97/">time_series (时间序列)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/model_retraining-%E6%A8%A1%E5%9E%8B%E9%87%8D%E8%AE%AD%E7%BB%83/">model_retraining (模型重训练)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/learnable_function_class/Sat, 18 Jul 2026 11:23:53 +0000https://terms-en.ai-term-hub.com/zh/terms/learnable_function_class/<h2 id="definition">Definition</h2> <p>在统计学习理论中，可学习函数类代表了算法可用的假设空间。它定义了模型潜在能够捕捉的模式或映射的范围，决定了模型的表达能力和泛化潜力。</p> <h3 id="summary">Summary</h3> <p>可学习函数类是由特定模型架构和参数空间定义的一组数学函数，学习算法可以对其进行优化。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>假设空间</li> <li>模型容量</li> <li>VC维</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>选择合适的模型架构</li> <li>泛化界限的理论分析</li> <li>理解模型的局限性</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bias-variance-tradeoff-%E5%81%8F%E5%B7%AE-%E6%96%B9%E5%B7%AE%E6%9D%83%E8%A1%A1/">Bias-variance tradeoff (偏差-方差权衡)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/model-capacity-%E6%A8%A1%E5%9E%8B%E5%AE%B9%E9%87%8F/">Model capacity (模型容量)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/hypothesis-space-%E5%81%87%E8%AE%BE%E7%A9%BA%E9%97%B4/">Hypothesis space (假设空间)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/kernel_density_estimation/Sat, 18 Jul 2026 11:22:59 +0000https://terms-en.ai-term-hub.com/zh/terms/kernel_density_estimation/<h2 id="definition">Definition</h2> <p>核密度估计（KDE）是一种基本的统计技术，通过对离散数据点进行平滑处理，生成连续的概率分布曲线。它在每个数据点处放置一个核函数（通常为高斯核），并将这些核函数叠加起来，从而估计出潜在的概率密度函数。</p> <h3 id="summary">Summary</h3> <p>一种非参数方法，用于根据有限的数据样本来估计随机变量的概率密度函数。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>概率密度函数</li> <li>非参数统计</li> <li>平滑处理</li> <li>高斯核</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>探索性数据分析 (EDA)</li> <li>单变量数据中的异常检测</li> <li>可视化数据集中的特征分布</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> scipy.stats <span style="color:#f92672">import</span> gaussian_kde </span></span><span style="display:flex;"><span><span style="color:#f92672">import</span> numpy <span style="color:#66d9ef">as</span> np </span></span><span style="display:flex;"><span> </span></span><span style="display:flex;"><span>data <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>random<span style="color:#f92672">.</span>normal(<span style="color:#ae81ff">0</span>, <span style="color:#ae81ff">1</span>, <span style="color:#ae81ff">100</span>) </span></span><span style="display:flex;"><span>kde <span style="color:#f92672">=</span> gaussian_kde(data) </span></span><span style="display:flex;"><span>x_vals <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>linspace(<span style="color:#f92672">-</span><span style="color:#ae81ff">3</span>, <span style="color:#ae81ff">3</span>, <span style="color:#ae81ff">100</span>) </span></span><span style="display:flex;"><span>y_vals <span style="color:#f92672">=</span> kde(x_vals) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%9B%B4%E6%96%B9%E5%9B%BE-histogram/">直方图 (Histogram)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%B8%95%E5%B0%94%E6%B3%BD%E7%AA%97-parzen-window/">帕尔泽窗 (Parzen Window)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%B8%A6%E5%AE%BD%E9%80%89%E6%8B%A9-bandwidth-selection/">带宽选择 (Bandwidth Selection)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/scipy%E7%BB%9F%E8%AE%A1%E6%A8%A1%E5%9D%97-scipy-stats/">SciPy统计模块 (Scipy Stats)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/glm/Sat, 18 Jul 2026 11:19:07 +0000https://terms-en.ai-term-hub.com/zh/terms/glm/<h2 id="definition">Definition</h2> <p>在统计建模中，GLM 代表广义线性模型（Generalized Linear Models），它将线性回归扩展到允许响应变量具有正态分布以外的误差分布模型。在人工智能领域，GLM 也常指代智谱 AI 推出的通用语言模型系列。</p> <h3 id="summary">Summary</h3> <p>GLM 通常指统计学中的广义线性模型，或智谱 AI 开发的通用语言模型系列。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>广义线性模型</li> <li>大语言模型</li> <li>前缀语言模型</li> <li>双向编码</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>统计预测</li> <li>自然语言处理</li> <li>文本生成</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transformer-transformer%E6%9E%B6%E6%9E%84/">Transformer (Transformer架构)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bert-bert%E6%A8%A1%E5%9E%8B/">BERT (BERT模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/linear-regression-%E7%BA%BF%E6%80%A7%E5%9B%9E%E5%BD%92/">Linear Regression (线性回归)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/generalized_additive_model_for_location_scale_and_shape/Sat, 18 Jul 2026 11:18:16 +0000https://terms-en.ai-term-hub.com/zh/terms/generalized_additive_model_for_location_scale_and_shape/<h2 id="definition">Definition</h2> <p>与传统回归模型仅关注均值不同，GAMLSS 对整个分布进行建模，包括位置（均值/中位数）、尺度（方差）、偏度和峰度。它使用广义&hellip;</p> <h3 id="summary">Summary</h3> <p>GAMLSS 是一种灵活的统计框架，允许使用加性平滑函数将所有分布参数建模为解释变量的函数。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>分布建模</li> <li>加性平滑</li> <li>方差建模</li> <li>统计灵活性</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>生长曲线分析</li> <li>风险评估</li> <li>复杂数据分布拟合</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/glm-%E5%B9%BF%E4%B9%89%E7%BA%BF%E6%80%A7%E6%A8%A1%E5%9E%8B/">GLM (广义线性模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/smoothing-splines-%E5%B9%B3%E6%BB%91%E6%A0%B7%E6%9D%A1/">Smoothing Splines (平滑样条)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/statistical-modeling-%E7%BB%9F%E8%AE%A1%E5%BB%BA%E6%A8%A1/">Statistical Modeling (统计建模)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/regression-analysis-%E5%9B%9E%E5%BD%92%E5%88%86%E6%9E%90/">Regression Analysis (回归分析)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/feature_scaling/Sat, 18 Jul 2026 11:17:13 +0000https://terms-en.ai-term-hub.com/zh/terms/feature_scaling/<h2 id="definition">Definition</h2> <p>特征缩放通过标准化输入变量的范围，防止量级较大的特征主导学习过程。常见方法包括归一化（最小-最大缩放）和标准化（Z-score标准化）。这一预处理步骤对于基于距离的算法和梯度下降优化至关重要，能加速收敛并提高模型性能。</p> <h3 id="summary">Summary</h3> <p>将数据的独立变量或特征的范围进行归一化，以确保量级一致性的过程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>归一化</li> <li>标准化</li> <li>梯度下降</li> <li>数据预处理</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>神经网络训练</li> <li>K-means聚类</li> <li>支持向量机 (SVM)</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> sklearn.preprocessing <span style="color:#f92672">import</span> StandardScaler </span></span><span style="display:flex;"><span><span style="color:#f92672">import</span> numpy <span style="color:#66d9ef">as</span> np </span></span><span style="display:flex;"><span> </span></span><span style="display:flex;"><span>X <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>array([[<span style="color:#ae81ff">1</span>, <span style="color:#ae81ff">2</span>], [<span style="color:#ae81ff">3</span>, <span style="color:#ae81ff">4</span>], [<span style="color:#ae81ff">5</span>, <span style="color:#ae81ff">6</span>]]) </span></span><span style="display:flex;"><span>scaler <span style="color:#f92672">=</span> StandardScaler() </span></span><span style="display:flex;"><span>X_scaled <span style="color:#f92672">=</span> scaler<span style="color:#f92672">.</span>fit_transform(X) </span></span><span style="display:flex;"><span>print(X_scaled) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/min-max-scaling-%E6%9C%80%E5%B0%8F-%E6%9C%80%E5%A4%A7%E7%BC%A9%E6%94%BE/">Min-Max Scaling (最小-最大缩放)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/z-score-normalization-z-score%E6%A0%87%E5%87%86%E5%8C%96/">Z-score Normalization (Z-score标准化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/data-preprocessing-%E6%95%B0%E6%8D%AE%E9%A2%84%E5%A4%84%E7%90%86/">Data preprocessing (数据预处理)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/gradient-descent-%E6%A2%AF%E5%BA%A6%E4%B8%8B%E9%99%8D/">Gradient Descent (梯度下降)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/empirical_dynamic_modeling/Sat, 18 Jul 2026 11:16:12 +0000https://terms-en.ai-term-hub.com/zh/terms/empirical_dynamic_modeling/<h2 id="definition">Definition</h2> <p>经验动态建模（EDM）是一种分析非线性动力系统的框架，它利用观测数据而不假设特定的参数形式。该方法依赖于Takens嵌入定理，通过重建状态空间来揭示系统内部的因果结构和动态规律。</p> <h3 id="summary">Summary</h3> <p>一种非参数统计方法，用于从观测时间序列数据中重构复杂系统的动态特性。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>状态空间重构</li> <li>Takens定理</li> <li>非线性</li> <li>因果推断</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>生态种群预测</li> <li>神经科学信号分析</li> <li>金融市场波动性预测</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/time-series-analysis-%E6%97%B6%E9%97%B4%E5%BA%8F%E5%88%97%E5%88%86%E6%9E%90/">Time series analysis (时间序列分析)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/chaos-theory-%E6%B7%B7%E6%B2%8C%E7%90%86%E8%AE%BA/">Chaos theory (混沌理论)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/state-space-%E7%8A%B6%E6%80%81%E7%A9%BA%E9%97%B4/">State space (状态空间)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/causal-discovery-%E5%9B%A0%E6%9E%9C%E5%8F%91%E7%8E%B0/">Causal discovery (因果发现)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/em_algorithm_and_gmm_model/Sat, 18 Jul 2026 11:15:46 +0000https://terms-en.ai-term-hub.com/zh/terms/em_algorithm_and_gmm_model/<h2 id="definition">Definition</h2> <p>该术语指期望最大化（EM）算法与高斯混合模型（GMM）之间的协同关系。GMM假设所有数据点都是由多个分布混合生成的，而EM算法则通过迭代方式优化这些混合成分的参数（如均值、协方差和权重），从而实现对数据的最佳拟合。</p> <h3 id="summary">Summary</h3> <p>期望最大化（EM）算法是一种用于在包含隐变量的统计模型中寻找最大似然估计的迭代方法，常用于拟合高斯混合模型。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>隐变量</li> <li>最大似然估计</li> <li>迭代优化</li> <li>高斯分布</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>语音识别中的音素建模</li> <li>图像分割</li> <li>客户细分分析</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/k-means%E8%81%9A%E7%B1%BB-%E4%B8%80%E7%A7%8D%E5%9F%BA%E4%BA%8E%E8%B7%9D%E7%A6%BB%E7%9A%84%E5%88%92%E5%88%86%E8%81%9A%E7%B1%BB%E7%AE%97%E6%B3%95/">K-means聚类 (一种基于距离的划分聚类算法)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%9A%90%E9%A9%AC%E5%B0%94%E5%8F%AF%E5%A4%AB%E6%A8%A1%E5%9E%8B-%E7%94%A8%E4%BA%8E%E5%BA%8F%E5%88%97%E6%95%B0%E6%8D%AE%E7%9A%84%E7%BB%9F%E8%AE%A1%E6%A8%A1%E5%9E%8B/">隐马尔可夫模型 (用于序列数据的统计模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E8%B4%9D%E5%8F%B6%E6%96%AF%E6%8E%A8%E6%96%AD-%E5%9F%BA%E4%BA%8E%E8%B4%9D%E5%8F%B6%E6%96%AF%E5%AE%9A%E7%90%86%E7%9A%84%E6%A6%82%E7%8E%87%E6%8E%A8%E7%90%86%E6%96%B9%E6%B3%95/">贝叶斯推断 (基于贝叶斯定理的概率推理方法)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%AF%86%E5%BA%A6%E4%BC%B0%E8%AE%A1-%E4%BC%B0%E8%AE%A1%E9%9A%8F%E6%9C%BA%E5%8F%98%E9%87%8F%E6%A6%82%E7%8E%87%E5%AF%86%E5%BA%A6%E5%87%BD%E6%95%B0%E7%9A%84%E8%BF%87%E7%A8%8B/">密度估计 (估计随机变量概率密度函数的过程)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/dataset_shift/Sat, 18 Jul 2026 11:12:42 +0000https://terms-en.ai-term-hub.com/zh/terms/dataset_shift/<h2 id="definition">Definition</h2> <p>当用于训练机器学习模型的数据分布与推理过程中遇到的数据分布不同时，就会发生数据集偏移。这种差异可能导致模型性能显著下降。</p> <h3 id="summary">Summary</h3> <p>数据集偏移是指训练数据与部署数据的输入统计特性发生变化的现象。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>协变量偏移</li> <li>概念漂移</li> <li>领域自适应</li> <li>泛化能力</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>监控生产环境中的机器学习模型</li> <li>重训练策略制定</li> <li>鲁棒性测试</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E8%BF%87%E6%8B%9F%E5%90%88-overfitting/">过拟合 (Overfitting)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%AC%A0%E6%8B%9F%E5%90%88-underfitting/">欠拟合 (Underfitting)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%A2%86%E5%9F%9F%E8%87%AA%E9%80%82%E5%BA%94-domain-adaptation/">领域自适应 (Domain Adaptation)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%95%B0%E6%8D%AE%E6%BC%82%E7%A7%BB-data-drift/">数据漂移 (Data Drift)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/data_driven_model/Sat, 18 Jul 2026 11:12:36 +0000https://terms-en.ai-term-hub.com/zh/terms/data_driven_model/<h2 id="definition">Definition</h2> <p>数据驱动模型是一种人工智能系统类型，其行为和预测结果源于历史数据中识别出的模式，而不是由硬编码的规则或明确的数学公式定义。这类模型能够自动适应数据中的复杂关系。</p> <h3 id="summary">Summary</h3> <p>其参数和结构直接从经验数据中学习，而非通过显式编程计算的计算模型。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>经验学习</li> <li>模式识别</li> <li>统计推断</li> <li>黑盒模型</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>使用历史价格数据预测股票市场趋势</li> <li>从医学影像扫描中诊断疾病</li> <li>使用卫星观测数据预测天气模式</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/machine_learning-%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0/">machine_learning (机器学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/empirical_modeling-%E7%BB%8F%E9%AA%8C%E5%BB%BA%E6%A8%A1/">empirical_modeling (经验建模)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/supervised_learning-%E7%9B%91%E7%9D%A3%E5%AD%A6%E4%B9%A0/">supervised_learning (监督学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/neural_networks-%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C/">neural_networks (神经网络)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/data_science_and_predictive_analytics/Sat, 18 Jul 2026 11:12:24 +0000https://terms-en.ai-term-hub.com/zh/terms/data_science_and_predictive_analytics/<h2 id="definition">Definition</h2> <p>数据科学涉及从结构和非结构化数据中提取知识的跨学科过程，而预测分析则专注于利用历史数据来预测未来……</p> <h3 id="summary">Summary</h3> <p>该领域结合统计分析和机器学习，从数据中提取见解并预测未来事件或行为。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>统计建模</li> <li>预测</li> <li>模式识别</li> <li>决策支持</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>客户流失预测</li> <li>金融风险评估</li> <li>供应链中的需求预测</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/machine-learning-%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0/">Machine Learning (机器学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/business-intelligence-%E5%95%86%E4%B8%9A%E6%99%BA%E8%83%BD/">Business Intelligence (商业智能)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/statistics-%E7%BB%9F%E8%AE%A1%E5%AD%A6/">Statistics (统计学)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/big-data-%E5%A4%A7%E6%95%B0%E6%8D%AE/">Big Data (大数据)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/cross_validation/Sat, 18 Jul 2026 11:12:12 +0000https://terms-en.ai-term-hub.com/zh/terms/cross_validation/<h2 id="definition">Definition</h2> <p>交叉验证是一种用于估计机器学习模型性能的统计方法。最常见的形式是k折交叉验证，即将数据分为k个相等的部分。模型在k-1个部分上进行训练，并在剩余的一个部分上进行测试，此过程重复k次，每次使用不同的部分作为测试集。</p> <h3 id="summary">Summary</h3> <p>一种重采样程序，通过将数据划分为子集进行训练和测试，用于在有限数据样本上评估机器学习模型。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>K折划分</li> <li>模型泛化能力</li> <li>过拟合检测</li> <li>性能估计</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>超参数调优</li> <li>比较不同算法</li> <li>在小数据集上验证模型稳定性</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> sklearn.model_selection <span style="color:#f92672">import</span> cross_val_score </span></span><span style="display:flex;"><span>cv_scores <span style="color:#f92672">=</span> cross_val_score(model, X, y, cv<span style="color:#f92672">=</span><span style="color:#ae81ff">5</span>) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/train-test-split-%E8%AE%AD%E7%BB%83-%E6%B5%8B%E8%AF%95%E9%9B%86%E5%88%92%E5%88%86/">Train-Test Split (训练-测试集划分)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/leave-one-out-%E7%95%99%E4%B8%80%E6%B3%95/">Leave-One-Out (留一法)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bootstrap-%E8%87%AA%E5%8A%A9%E6%B3%95/">Bootstrap (自助法)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/category_utility/Sat, 18 Jul 2026 11:09:42 +0000https://terms-en.ai-term-hub.com/zh/terms/category_utility/<h2 id="definition">Definition</h2> <p>该指标量化了一组类别在多大程度上允许人们预测这些类别内属性的值。它在类别大小与其内容同质性之间取得平衡，是概念学习和聚类评估中的重要指标。</p> <h3 id="summary">Summary</h3> <p>类别效用是一种数学度量，用于根据分类方案提供的信息增益来评估分类方案的有效性。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>信息增益</li> <li>聚类评估</li> <li>预测准确性</li> <li>概念学习</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>评估聚类质量</li> <li>概念获取研究</li> <li>优化特征选择</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/entropy-%E7%86%B5/">entropy (熵)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mutual-information-%E4%BA%92%E4%BF%A1%E6%81%AF/">mutual information (互信息)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/clustering-%E8%81%9A%E7%B1%BB/">clustering (聚类)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/concept-formation-%E6%A6%82%E5%BF%B5%E5%BD%A2%E6%88%90/">concept formation (概念形成)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/bradleyterry_model/Sat, 18 Jul 2026 11:09:30 +0000https://terms-en.ai-term-hub.com/zh/terms/bradleyterry_model/<h2 id="definition">Definition</h2> <p>布拉德利-特里模型是一种概率模型，广泛用于心理测量学和机器学习中以处理成对比较。它为每个项目分配一个潜在分数，计算概率……</p> <h3 id="summary">Summary</h3> <p>一种用于分析成对比较数据的统计模型，估计一个项目优于另一个项目的概率。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>成对比较</li> <li>排名系统</li> <li>逻辑回归</li> <li>偏好学习</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>体育排名算法</li> <li>A/B测试分析</li> <li>基于人类反馈的强化学习 (RLHF)</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/elo-rating-elo-%E7%AD%89%E7%BA%A7%E5%88%86/">Elo Rating ( Elo 等级分)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/plackett-luce-model-%E6%99%AE%E6%8B%89%E5%88%87%E7%89%B9-%E5%8D%A2%E6%96%AF%E6%A8%A1%E5%9E%8B/">Plackett-Luce Model (普拉切特-卢斯模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/isotonic-regression-%E4%BF%9D%E5%BA%8F%E5%9B%9E%E5%BD%92/">Isotonic Regression (保序回归)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/rlhf-%E5%9F%BA%E4%BA%8E%E4%BA%BA%E7%B1%BB%E5%8F%8D%E9%A6%88%E7%9A%84%E5%BC%BA%E5%8C%96%E5%AD%A6%E4%B9%A0/">RLHF (基于人类反馈的强化学习)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/biasvariance_tradeoff/Sat, 18 Jul 2026 11:09:15 +0000https://terms-en.ai-term-hub.com/zh/terms/biasvariance_tradeoff/<h2 id="definition">Definition</h2> <p>偏差-方差权衡描述了欠拟合（高偏差）和过拟合（高方差）之间的张力。高偏差模型对数据做出强烈假设，可能会忽略相关关系；而高方差模型则过于关注训练数据中的噪声。</p> <h3 id="summary">Summary</h3> <p>监督学习中的一个基本问题，最小化误差需要在模型复杂度和泛化能力之间取得平衡。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>欠拟合</li> <li>过拟合</li> <li>泛化</li> <li>模型复杂度</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>模型选择</li> <li>超参数调优</li> <li>公平性审计</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/regularization-%E6%AD%A3%E5%88%99%E5%8C%96/">Regularization (正则化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/cross-validation-%E4%BA%A4%E5%8F%89%E9%AA%8C%E8%AF%81/">Cross-Validation (交叉验证)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/ensemble-methods-%E9%9B%86%E6%88%90%E6%96%B9%E6%B3%95/">Ensemble Methods (集成方法)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/overfitting-%E8%BF%87%E6%8B%9F%E5%90%88/">Overfitting (过拟合)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/base_rate/Sat, 18 Jul 2026 11:08:27 +0000https://terms-en.ai-term-hub.com/zh/terms/base_rate/<h2 id="definition">Definition</h2> <p>在统计学和机器学习中，基础比率指的是给定数据集中某种条件或结果的潜在频率。忽视基础比率往往会导致基础比率谬误，即人们倾向于忽略先验概率而过度依赖具体证据。</p> <h3 id="summary">Summary</h3> <p>基础比率是指在一个群体中某事件发生的先验概率，独立于任何特定证据或测试结果。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>先验概率</li> <li>类别不平衡</li> <li>贝叶斯推断</li> <li>基础比率谬误</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>医学诊断</li> <li>欺诈检测</li> <li>风险评估</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/precision-%E7%B2%BE%E7%A1%AE%E7%8E%87/">Precision (精确率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/recall-%E5%8F%AC%E5%9B%9E%E7%8E%87/">Recall (召回率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bayes-theorem-%E8%B4%9D%E5%8F%B6%E6%96%AF%E5%AE%9A%E7%90%86/">Bayes&rsquo; Theorem (贝叶斯定理)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/astrostatistics/Sat, 18 Jul 2026 11:07:14 +0000https://terms-en.ai-term-hub.com/zh/terms/astrostatistics/<h2 id="definition">Definition</h2> <p>天体统计学是连接统计学与天文学的专业领域。它涉及开发和应用严谨的统计技术，以应对天文数据带来的独特挑战，例如处理海量观测数据、噪声建模以及从稀疏或不完整的数据中提取物理参数。</p> <h3 id="summary">Summary</h3> <p>应用统计方法分析天文数据并解决天体物理学问题的学科。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>贝叶斯推断</li> <li>数据挖掘</li> <li>误差分析</li> <li>宇宙学参数估计</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>分析星系分布模式</li> <li>估算暗物质特性</li> <li>处理引力波探测器的信号</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/machine-learning-%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0/">Machine Learning (机器学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/data-science-%E6%95%B0%E6%8D%AE%E7%A7%91%E5%AD%A6/">Data Science (数据科学)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/cosmology-%E5%AE%87%E5%AE%99%E5%AD%A6/">Cosmology (宇宙学)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/signal-processing-%E4%BF%A1%E5%8F%B7%E5%A4%84%E7%90%86/">Signal Processing (信号处理)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/ab_testing/Sat, 18 Jul 2026 11:02:40 +0000https://terms-en.ai-term-hub.com/zh/terms/ab_testing/<h2 id="definition">Definition</h2> <p>A/B 测试是一种随机对照实验，通过比较两个变体（A 和 B）来评估哪一个在特定指标上能产生更好的结果。在人工智能工程中，它对优化至关重要。</p> <h3 id="summary">Summary</h3> <p>一种统计方法，通过比较变量的两个版本来确定哪个表现更好。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>对照组</li> <li>统计显著性</li> <li>假设检验</li> <li>随机化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>优化推荐引擎的点击率</li> <li>比较不同模型架构的准确性</li> <li>测试 AI 驱动应用中的用户界面更改</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/multivariate-testing-%E5%A4%9A%E5%8F%98%E9%87%8F%E6%B5%8B%E8%AF%95/">Multivariate Testing (多变量测试)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/statistical-power-%E7%BB%9F%E8%AE%A1%E5%8A%9F%E6%95%88/">Statistical Power (统计功效)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/conversion-rate-optimization-%E8%BD%AC%E5%8C%96%E7%8E%87%E4%BC%98%E5%8C%96/">Conversion Rate Optimization (转化率优化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/randomized-controlled-trial-%E9%9A%8F%E6%9C%BA%E5%AF%B9%E7%85%A7%E8%AF%95%E9%AA%8C/">Randomized Controlled Trial (随机对照试验)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/prior/Sat, 18 Jul 2026 10:53:51 +0000https://terms-en.ai-term-hub.com/zh/terms/prior/<h2 id="definition">Definition</h2> <p>“先验”代表了在纳入新观测值之前，关于某个变量的现有信念或历史数据。在贝叶斯推断中，先验与观测数据的似然度相结合，以更新对该变量的认知，形成后验分布。</p> <h3 id="summary">Summary</h3> <p>在贝叶斯统计中，表示在观察新证据或数据之前，对某个参数的知识或信念的概率分布。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>贝叶斯推断</li> <li>概率分布</li> <li>正则化</li> <li>后验</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>实现贝叶斯神经网络</li> <li>应用L1/L2正则化作为先验约束</li> <li>结合新的测试结果更新医学诊断</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/posterior-%E5%90%8E%E9%AA%8C/">Posterior (后验)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/likelihood-%E4%BC%BC%E7%84%B6/">Likelihood (似然)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bayesian-%E8%B4%9D%E5%8F%B6%E6%96%AF/">Bayesian (贝叶斯)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/probability-%E6%A6%82%E7%8E%87/">Probability (概率)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/monte/Sat, 18 Jul 2026 10:53:03 +0000https://terms-en.ai-term-hub.com/zh/terms/monte/<h2 id="definition">Definition</h2> <p>蒙特卡洛技术是一类依赖重复随机抽样来估算数学量的计算算法。它们在多维积分、概率推断以及处理具有不确定性的复杂系统时特别有用，通过大量样本的平均值来逼近真实解。</p> <h3 id="summary">Summary</h3> <p>蒙特卡洛指代蒙特卡洛方法，该方法使用重复的随机抽样来获得原则上确定但过于复杂而无法解析求解的问题的数值结果。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>随机抽样</li> <li>数值近似</li> <li>概率推断</li> <li>大数定律</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>贝叶斯后验估计</li> <li>不确定性下的机器人路径规划</li> <li>金融期权定价</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mcmc-%E9%A9%AC%E5%B0%94%E5%8F%AF%E5%A4%AB%E9%93%BE%E8%92%99%E7%89%B9%E5%8D%A1%E6%B4%9B/">MCMC (马尔可夫链蒙特卡洛)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/importance-sampling-%E9%87%8D%E8%A6%81%E6%80%A7%E9%87%87%E6%A0%B7/">Importance Sampling (重要性采样)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bootstrap-%E8%87%AA%E5%8A%A9%E6%B3%95/">Bootstrap (自助法)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/simulation-%E6%A8%A1%E6%8B%9F/">Simulation (模拟)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/gaussian/Sat, 18 Jul 2026 10:51:30 +0000https://terms-en.ai-term-hub.com/zh/terms/gaussian/<h2 id="definition">Definition</h2> <p>高斯指的是正态分布，这是一种由均值和方差表征的连续概率分布。在AI中，它广泛用于概率建模、贝叶斯推断以及</p> <h3 id="summary">Summary</h3> <p>与正态分布相关，这是一种钟形曲线，是统计学和AI噪声建模的基础。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>正态分布</li> <li>均值</li> <li>方差</li> <li>概率密度</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>噪声建模</li> <li>贝叶斯网络</li> <li>权重初始化</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">import</span> numpy <span style="color:#66d9ef">as</span> np </span></span><span style="display:flex;"><span><span style="color:#75715e"># Generate 1000 samples from a standard Gaussian distribution</span> </span></span><span style="display:flex;"><span>samples <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>random<span style="color:#f92672">.</span>normal(loc<span style="color:#f92672">=</span><span style="color:#ae81ff">0.0</span>, scale<span style="color:#f92672">=</span><span style="color:#ae81ff">1.0</span>, size<span style="color:#f92672">=</span><span style="color:#ae81ff">1000</span>) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/distribution-%E5%88%86%E5%B8%83/">Distribution (分布)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bell-curve-%E9%92%9F%E5%BD%A2%E6%9B%B2%E7%BA%BF/">Bell Curve (钟形曲线)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/standard-deviation-%E6%A0%87%E5%87%86%E5%B7%AE/">Standard Deviation (标准差)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/causal/Sat, 18 Jul 2026 10:49:42 +0000https://terms-en.ai-term-hub.com/zh/terms/causal/<h2 id="definition">Definition</h2> <p>在人工智能中，因果建模旨在理解对某一变量的干预如何影响另一变量。与依赖观察模式的预测模型不同，因果AI使用结构方程模型等工具来揭示变量间的内在机制。</p> <h3 id="summary">Summary</h3> <p>因果推断涉及确定变量之间的因果关系，而不仅仅是识别统计相关性。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>干预</li> <li>反事实推理</li> <li>有向无环图</li> <li>混杂变量</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>医疗疗效分析</li> <li>经济政策影响评估</li> <li>工业系统中的根本原因分析</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/correlation-%E7%9B%B8%E5%85%B3%E6%80%A7/">Correlation (相关性)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/inference-%E6%8E%A8%E6%96%AD/">Inference (推断)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/structural-equation-modeling-%E7%BB%93%E6%9E%84%E6%96%B9%E7%A8%8B%E5%BB%BA%E6%A8%A1/">Structural Equation Modeling (结构方程建模)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/do-calculus-do%E6%BC%94%E7%AE%97/">Do-Calculus (do演算)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/carlo/Sat, 18 Jul 2026 10:49:30 +0000https://terms-en.ai-term-hub.com/zh/terms/carlo/<h2 id="definition">Definition</h2> <p>蒙特卡洛方法是AI和统计学中用于近似难以解析求解的复杂数学问题的关键技术。通过生成成千上万个随机样本来估算结果。</p> <h3 id="summary">Summary</h3> <p>指蒙特卡洛方法，这是一类依赖重复随机采样以获得数值结果的计算算法。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>随机采样</li> <li>统计近似</li> <li>模拟</li> <li>概率估计</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>通过模拟估计强化学习中的状态值。</li> <li>使用马尔可夫链蒙特卡洛（MCMC）进行贝叶斯后验推断。</li> <li>为高维空间中的概率模型计算积分。</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">import</span> numpy <span style="color:#66d9ef">as</span> np </span></span><span style="display:flex;"><span><span style="color:#75715e"># Monte Carlo estimation of Pi</span> </span></span><span style="display:flex;"><span><span style="color:#66d9ef">def</span> <span style="color:#a6e22e">estimate_pi</span>(samples): </span></span><span style="display:flex;"><span> points <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>random<span style="color:#f92672">.</span>uniform(<span style="color:#f92672">-</span><span style="color:#ae81ff">1</span>, <span style="color:#ae81ff">1</span>, size<span style="color:#f92672">=</span>(samples, <span style="color:#ae81ff">2</span>)) </span></span><span style="display:flex;"><span> inside <span style="color:#f92672">=</span> np<span style="color:#f92672">.</span>sum(points[:, <span style="color:#ae81ff">0</span>]<span style="color:#f92672">**</span><span style="color:#ae81ff">2</span> <span style="color:#f92672">+</span> points[:, <span style="color:#ae81ff">1</span>]<span style="color:#f92672">**</span><span style="color:#ae81ff">2</span> <span style="color:#f92672">&lt;=</span> <span style="color:#ae81ff">1</span>) </span></span><span style="display:flex;"><span> <span style="color:#66d9ef">return</span> <span style="color:#ae81ff">4</span> <span style="color:#f92672">*</span> inside <span style="color:#f92672">/</span> samples </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/monte_carlo-%E8%92%99%E7%89%B9%E5%8D%A1%E6%B4%9B/">Monte_Carlo (蒙特卡洛)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/simulation-%E6%A8%A1%E6%8B%9F/">simulation (模拟)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/random_sampling-%E9%9A%8F%E6%9C%BA%E9%87%87%E6%A0%B7/">random_sampling (随机采样)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mcmc-%E9%A9%AC%E5%B0%94%E5%8F%AF%E5%A4%AB%E9%93%BE%E8%92%99%E7%89%B9%E5%8D%A1%E6%B4%9B/">MCMC (马尔可夫链蒙特卡洛)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/bayesian/Sat, 18 Jul 2026 10:49:18 +0000https://terms-en.ai-term-hub.com/zh/terms/bayesian/<h2 id="definition">Definition</h2> <p>AI中的贝叶斯方法利用概率论，随着更多证据的出现来更新假设的可能性。这种方法允许模型量化不确定性并动态优化预测结果。</p> <h3 id="summary">Summary</h3> <p>指基于贝叶斯定理的统计方法，用于根据新证据更新概率。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>贝叶斯定理</li> <li>先验概率</li> <li>后验概率</li> <li>不确定性量化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>垃圾邮件过滤</li> <li>医疗诊断系统</li> <li>A/B测试分析</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> sklearn.naive_bayes <span style="color:#f92672">import</span> GaussianNB </span></span><span style="display:flex;"><span>model <span style="color:#f92672">=</span> GaussianNB() </span></span><span style="display:flex;"><span>model<span style="color:#f92672">.</span>fit(X_train, y_train) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/probability-%E6%A6%82%E7%8E%87/">Probability (概率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/naive-bayes-%E6%9C%B4%E7%B4%A0%E8%B4%9D%E5%8F%B6%E6%96%AF/">Naive Bayes (朴素贝叶斯)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/inference-%E6%8E%A8%E6%96%AD/">Inference (推断)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/statistics-%E7%BB%9F%E8%AE%A1%E5%AD%A6/">Statistics (统计学)</a></li> </ul>