统计学习理论的本质:英中文术语对照表
来源:张学工译, VN Vapnik原著, 统计学习理论的本质, 清华大学出版社, 2000
使用范围:南京师范大学计算机科学与技术学院研究生。
声明:任何人在其出版物使用或者上载到互连网都必须得到译者及出版社的许可。
AdaBoost algorithm (AdaBoost(自举)算法)163
admissible structure (容许结构) 95
algorithmic complexity (算法复杂度) 10
annealed entropy (退火熵) 55
ANOVA decomposition (ANOVA分解) 199
a posteriori information (后验信息) 120
a priori information (先验信息) 120
approximately defined operator (近似定义的算子) 230 approximation rate (逼近速率) 98
artificial intelligence (人工智能) 13
axioms of probability theory (概率理论的公理) 60
back propagation method (后向传播方法) 126传统美食作文
basic problem of probability theory (概率论的基本问题) 62
basic problem of statistics (统计学的基本问题) 63
Bayesian approach (贝叶斯方法) 119
Bayesian inference (贝叶斯推理) 34
bound on the distance to the smallest risk (与最小风险的距离的界) 77 bound on the values of achieved risk (所得风险值的界) 77
bounds on generalization ability of a learning machine (学习机器推广能力的界) 76
canonical parating hyperplanes (标准分类超平面) 132
capacity control problem (容量控制问题) 116
cau-effect relation (因果关系) 9
choosing the best spar algebraic polynomial (选择最佳稀疏多项式)117
choosing the degree of polynomial (选择多项式阶数) 116 classification error (分类错误) 19
codebook (码本) 106
complete (Popper's) nonfalsifiability (完全(波普)不可证伪性) 52 compression coefficient (压缩系数) 107
consistency of inference (推理的一致性) 36
constructive distribution-independent bound on the rate of convergence (构造性的不依赖于分布的收敛速度界) 69
convolution of inner production (内积回旋) 140
criterion of nonfalsifiability (不可证伪性判据) 47
data smoothing problem (数据平滑问题) 209
decision-making problem (决策选择问题) 296
decision trees (决策树) 7
deductive inference (演绎推理) 47
density estimation problem (密度估计问题):
parametric(Fisher-Wald) tting(参数化(Fisher-Wald)表示) 20
nonparametric tting (非参数表示) 28
discrepancy (差异) 18
discriminant analysis (判别分析) 24
discriminant function (判别函数) 25
distribution-dependent bound on the rate of convergence (依赖于分布的收敛速度界) 69
distribution-independent bound on the rate of convergence (不依赖于分布的收敛速度界) 69
Δ
Δ-margin parating hyperplane (间隔分类超平面) 132 empirical distribution function (经验分布函数) 28
empirical process (经验过程) 40
empirical risk functional (经验风险泛函) 20
empirical risk minimization inductive principle (经验风险最小化归纳原则) 20
enmble of support vector machines (支持向量机的组合) 163 entropy of the t of functions (函数集的熵) 42
entropy on the t of indicator functions (指示函数集的熵) 42 equivalence class (等价类) 292
estimation of the values of a function at the given points (估计函数在给定点上的值) 292
expert systems (专家系统) 7
ε-innsitivity (ε不敏感性) 181
ε-innsitive loss function (ε不敏感损失函数) 181
feature lection problem (特征选择问题) 118
function approximation (函数逼近) 98
function estimation model (函数估计模型) 17
Gaussian (高斯函数) 26
generalized Glivenko-Cantelli problem (广义Glivenko-Cantelli问题)66
generalized growth function (广义生长函数) 85
generator random vectors (随机向量产生器) 17
Glivenko-Cantelli problem (Glivenko-Cantelli问题) 66
growth function (生长函数) 55
Hamming distance (汉明距离) 104
handwritten digit recognition (手写数字识别) 146
hard threshold vicinity function (硬限邻域函数) 103
hard vicinity function (硬领域函数) 269
hidden Markov models (隐马尔可夫模型) 7
hidden units (隐结点) 101
Huber loss function (Huber损失函数) 183
ill-pod problems (不适定问题): 9
solution by variation method (变分方法解) 236
solution by residual method (残差方法解) 236
solution by quasi-solution method (拟解方法解) 236 independent trials (独立试验) 62
inductive inference (归纳推理) 50
inner product in Hilbert space (希尔伯特空间中的内积) 140 integral equations (积分方程):
solution for exact determined equations (精确确定的方程的解)237
solution for approximately determined equations (近似确定的方程的解) 237
kernel function (核函数) 27
Kolmogorov-Smirnov distribution (Kolmogorov-Smirnov分布) 87 Kulback-Leibler distance (Kulback-Leibler距离) 32
Kuhn-Tücker conditions (库恩-塔克条件) 134
Lagrangian multiplier (拉格朗日乘子) 133
Lagrangian (拉格朗日函数) 133
Laplacian (拉普拉斯函数) 277
law of large number in the functional space (泛函空间中的大数定律)41
law of large numbers (大数定律) 39
law of large numbers in vector space (向量空间中的大数定律) 41 Lie derivatives (Lie导数) 20
learning matrices (学习矩阵) 7
least-squares method (最小二乘方法) 21
least-modulo method (最小模方法) 182
linear discriminant function (学习判别函数) 31
linearly nonparable ca (线性不可分情况) 135
local approximation (局部逼近) 104
local risk minimization (局部风险最小化) 103
locality parameter (局部性参数) 103
loss-function (损失函数):
for AdaBoost algorithm (AdaBoost算法的损失函数) 163油卡管理
for density estimation (密度估计的损失函数) 21
for logistic regression (逻辑回归的损失函数) 156
for pattern recognition (模式识别的损失函数) 21
for regression estimation (回归估计的损失函数) 21 madaline(Madaline自适应学习机) 7
main principle for small sample size problems (小样本数问题的基本原则) 28
maximal margin hyperplane (最大间隔超平面) 131
maximum likehood method (最大似然方法) 24
McCulloch-Pitts neuron model (McCulloch-Pitts神经元模型) 2 measurements with the additive noi (加性噪声下的测量) 25 metric ε-entropy (ε熵度量) 44
沈德潜minimum description length principle (最小描述长度原则) 104 mixture of normal densities (正态密度的组合) 26
National Institute of Standard and Technology (NIST) digit databa (美国国家标准技术研究所(NIST)数字数据库) 173
neural networks (神经网络) 126
non-trivially consistent inference (非平凡一致推理) 36 nonparametric density estimation (非参数密度估计) 27
normal discriminant function (正态判别函数) 31
one-sided empirical process (单边经验过程) 40
optimal parating hyperplane (最优分类超平面) 131
overfitting phenomenon (过学习现象) 14
parametric methods of density estimation (密度估计的参数方法) 24 partial nonfalsifiability (部分不可证伪性) 51
Parzen's windows method (Parzen窗方法) 27
pattern recognition problem (模式识别问题) 19
perceptron (感知器) 1
perceptron's stopping rule (感知器迭代终止规则) 6
polynomial approximation of regression (回归的多项式逼近) 116 polynomial machine (多项式机器) 143
potential nonfalsifiability (潜在不可证伪性) 53
probability measure (概率测度) 59
probably approximately correct (PAC) model (可能近似正确(PAC)模型) 13
problem of demarcation (区分问题) 49
pudo-dimension (伪维) 90
quadratic programming problem (二次规划问题) 133
quantization of parameters (参数的量化) 110
quasi-solution (拟解) 112
radial basis function machine (径向基函数机器) 144
random entropy (随机熵) 42
radnom string (随机串) 10
randomness concept (随机性概念) 10
regression estimation problem (回归估计问题) 19
regression function (回归函数) 19
regularization theory (正则化理论) 9
regularized functional (正则化泛函) 9
reproducing kernel Hilbert space (再生核希尔伯特空间) 244 residual principle (残差原则) 236
rigorous (distribution-dependent) bounds (严格(依赖于分布的)界) 85 risk functional (风险泛函) 18
risk minimization from empirical data problem (基于经验数据最小化风险的问题) 20
robust estimators (鲁棒估计) 26
robust regression (鲁棒回归) 26
Ronblatt's algorithm (Ronblatt算法) 5
t of indicators (指示器集合) 73
t of unbounded functions (无界函数集合) 77
σ-algebra (σ代数) 60
sigmoid function (S型(sigmoid)函数) 125
small samples size (小样本数) 93
smoothing kernel (平滑核) 100花雕醉鸡
smoothness of functions (函数的平滑性) 100
soft threshold vicinity function (软阈值领域函数) 103
soft vicinity function (软领域函数) 269
soft-margin parating hyperplane (软间隔分类超平面) 135
spline function (样条函数):
with a finite number of nodes (有限结点的样条函数) 194
with an infinite number of nodes (无穷多结点的样条函数) 195 stochastic approximation stopping rule (随机逼近终止规则) 34 stochastic ill-pod problems (随机不适定问题) 113
strong mode estimating a probability measure (强方式概率度量估计)63
structural risk minimization principle (结构风险最小化原则) 94 structure (结构) 94两拼音节有哪些
structure of growth function (生长函数的结构) 79
supervisor (训练器) 17
support vector machines (支持向量机) 137
support vectors (支持向量) 134
support vector ANOVA decomposition (支持向量ANOVA分解) 199 SVM n approximation of the logistic regression (逻辑回归的SVM n逼近) 155
SVM density estimator (SVM密度估计) 246
SVM conditional probability estimator (SVM条件概率估计) 257 tails of distribution (分布的尾部) 78
tangent distance (切距) 149
training t (训练集) 18
transductive inference (转导推理) 293
Turing-Church thesis (Turing-Church理论) 177
日本全面侵华two layer neural networks machine (两层神经网络机器) 145
two-sided empirical process (双边经验过程) 40
U.S. Postal Service digit databa (美国邮政数字数据库) 173 uniform one-sided convergence (一致单边收敛) 39
uniform two-sided convergence (一致双边收敛) 39
球宝
VC dimension of a t of indictor functions (指示函数集的VC维) 79 VC dimension of a t of real functions (实函数集的VC维) 81
VC entropy (VC熵) 44
VC subgraph (VC子图) 90
管理创新案例
vicinal risk minimization method(领域风险最小化) 268
vicinity kernel(领域核):273
one-vicinal kernel (单领域核) 273
two-vicinal kernel (双领域核) 273
VRM method (VRM方法):