Absolute value rectification
Activation Function
Accumulated error backpropagation
Acoustic modeling
Acquisition funtion
Actor-critic method
Adaptive bitrate (ABR) algorithm
Adaptive Resonance Theory/ART
Addictive model
Adversarial example
Adversarial Networks
Affine Layer
Affinity matrix
Agent
Algorithm
Alpha-beta pruning
Alternative splicing dataset
Analytic gradient
Ancestral Sampling
Annealed importance sampling
Anomaly detection
Absolute value rectification
Activation Function
Accumulated error backpropagation
Acoustic modeling
Acquisition funtion
Actor-critic method
Adaptive bitrate (ABR) algorithm
Adaptive Resonance Theory/ART
Addictive model
Adversarial example
Adversarial Networks
Affine Layer
Affinity matrix
Agent
Algorithm
Alpha-beta pruning
Alternative splicing dataset
Analytic gradient
Ancestral Sampling
Annealed importance sampling
Anomaly detection
Application-specific integrated circuit
Approximate Bayesian computation
Approximate inference
Approximation
Architecture
Area Under ROC Curve/AUC
Artificial General Intelligence/AGI
Artificial Intelligence/AI
Association analysis
Asymptotically unbiased
Asynchoronous Stochastic Gradient Descent
Attention mechanism
Attribute conditional independence assumption 
Attribute space
Attribute value
Augmented Lagrangian
Autoencoder
Automatic differentiation
Automatic speech recognition/ASR
Automatic summarization
Auto-regressive network
Average gradient
Average-Pooling
Backpropagation/BP
Backpropagation Through Time
Backward induction
Bag of words/BoW
Base learner
Base learning algorithm
Batch
Batch Normalization/BN
Bayes decision rule 
Bayes error
Bayes Model Averaging/BMA
Bayes optimal classifier
Bayesian decision theory
Bayesian network
Bayesian optimization
Beam search
Bechmark
Belief network
Bellman equation
Between-class scatter matrix
Bias
Biased
Biased importance sampling
Bias-variance decomposition
Bias-Variance Dilemma
Bi-directional Long-Short Term Memory/Bi-LSTM
Binary classification
Binary relation
Binary sparse coding
Binomial distribution
Binomial test
Bi-partition
Block coordinate descent
Block Gibbs Sampling
Boilerplate code
Boltzmann distribution
Boltzmann machine
Bootstrap sampling
Bootstrapping
Bottleneck layer
Bounding Boxes
Break-Event Point/BEP
Bridge sampling
Broadcasting
Burning-in
Calculus of variations
Calibration
Canonical
Cascade/Coalesced
Cascade-Correlation
Categorical attribute
Categorical distribution
Causal factor
Causal modeling
Centered difference
Central limit theorem
Chain rule
Chordal graph
Class-conditional probability
Classification and regression tree/CART
Classifier
Class-imbalance
Clip gradient
Clique potential
Closed-form
Cluster
Cluster analysis
Clustering
Clustering ensemble
Co-adapting
Coding matrix
Collaborative filtering
COLT
Committee-based learning	
Competitive learning	
Complete graph
Component learner	
Comprehensibility	
Computation Cost
Computational Linguistics
Computer vision
Concept drift
Concept Learning System/CLS
Conditional entropy
Conditional mutual information
Conditional Probability Table/CPT
Conditional random field/CRF
Conditional risk 
Confidence
Confusion matrix	
Conjugate directions
Conjugate distribution
Conjugate gradient
Connection weight	
Connectionism	
Consistency	
Consistency convergence
Contingency table	
Continuation method
Continuous attribute	
Contractive autoencoder
Contractive neural network
Convex optimization
Convergence
Conversational agent	
Convex quadratic programming 	
Convexity
Convolutional Boltzmann Machine
Convolutional neural network/CNN
Co-occurrence	
Coordinate descent
Correlation coefficient	
Cosine similarity
Cost curve	
Cost Function	
Cost matrix	
Cost-sensitive	
Covariance
Covariance matrix
Cross entropy
Cross validation	
Cross-correlation
Crowdsourcing	
Cumulative function
Curse of dimensionality
Curve-fitting
Cut point	
Cutting plane algorithm 
Data generating distribution
Data mining	
Data parallelism
Data set	
Data Wrangling
Dataset augmentation
Debugging strategy
Decision Boundary	
Decision stump
Decision tree	
Deconvolutional Network
Deduction	
Deep Belief Network	
Deep Boltzmann Machine
Deep circuit
Deep Convolutional Generative Adversarial Network/DCGAN
Deep generative model
Deep learning
Deep neural network/DNN
Deep Q-Learning	
Deep Q-Network	
Denoising autoencoder
Denoising score matching
Density estimation 	
Density-based clustering
Detailed balance
Determinant
Deterministic
Diagonal matrix
Differentiable neural computer
Differential entropy
Differential equation
Dimensionality reduction algorithm
Directed edge
Directed graphical model
Directional derivative
Dirichlet distribution
Disagreement measure	
Discriminative model	
Discriminator
Discriminator network
Distance measure	
Distance metric learning
Distribution
Divergence
Diversity measure	
Domain adaption	
Dominant strategy
Double backprop
Doubly block circulant matrix
Downsampling	
D-separation/Directed separation
Dual problem 
Dummy node	
Dynamic Fusion	
Dynamic programming
Echo state network
Edge device
Eigendecomposition
Eigenvalue
Eigenvalue decomposition
Eigenvector
Element-wise product
Ellipsoid method
Embedding	
Emotional analysis	
Empirical conditional entropy
Empirical entropy
Empirical error
Empirical risk	
End-to-End	
Energy-based model	
Ensemble learning
Ensemble pruning	
Epochs
Error Correcting Output Codes/ECOC	
Error rate
Error-ambiguity decomposition
Euclidean distance
Euclidean norm
Evolutionary computation 
Exact
Expectation-Maximization/EM
Expected loss 
Expert network
Explaining away effect
Exploding Gradient Problem
Exploitation
Exploration
Exponential loss function 
Extreme Learning Machine/ELM
Factor analysis
Factorization
Factors of variation
False negative	
False positive
False Positive Rate/FPR	
Fault-tolerant asynchronous training
Feature engineering	
Feature extractor
Feature map
Feature selection
Feature vector
Featured Learning	
Feedforward Neural Networks/FNN
Field Programmable Gated Array
Fine-tuning	
Finite difference
Fixed point equation
Flipping output	
Fluctuation	
Folk Theorem
Forget gate
Forward stagewise algorithm
Fourier transform
Frequentist	
Frequentist probability
Full-rank matrix
Functional derivative
Functional neuron
Gain ratio
Game payoff
Game theory	
Gated recurrent net/GRN
Gaussian kernel function
Gaussian Mixture Model	
Gaussian Process
General Problem Solving	
Generalization
Generalization error
Generalization error bound
Generalized Lagrange function
Generalized linear model	
Generalized pseudolikelihood
Generalized Rayleigh quotient	
Generalized score matching	
Generative Adversarial Networks/GAN
Generative Model	
Generative moment matching network
Generator	
Genetic Algorithm/GA
Giant magnetoresistance
Gibbs sampling
Gini index
Global contrast normalization
Global minimum	
Global Optimization
Gradient boosting tree
Gradient Descent
Gradient energy distribution
Graph theory	
Grid search	
Ground-truth	
Hard  margin 
Hard voting
Harmonic mean	
Hesse matrix
Heterogeneous Information Network/HIN
Hidden dynamic model
Hidden layer
Hidden Markov Model/HMM
Hierarchical clustering
Hilbert space
Hinge loss function
Hold-out
Homogeneous
Hybrid computing
Hyperparameter
Hypothesis
Hypothesis test
ICML
Identity matrix
Image restoration
Imperfect Information
Improved iterative scaling/IIS
Incremental learning	
Independent and identically distributed/i.i.d.
Independent Component Analysis/ICA
Independent subspace analysis
Indicator function
Individual learner	
Induction	
Inductive bias	
Inductive learning
Inductive Logic Programming/ILP	
Inequality constraint
Inference
Information entropy	
Information gain
Input layer
Insensitive loss	
Instance segmentation	
Inter-cluster similarity 
International Conference for Machine Learning/ICML
Intra-cluster similarity
Intrinsic value	
Invariance
Invert
Isometric Mapping/Isomap
Isotonic regression	
Iterative Dichotomiser	
Jensen-Shannon Divergence/JSD
Kernel method
Kernel trick
Kernelized Linear Discriminant Analysis/KLDA 
K-fold cross validation	
K-Means Clustering	
K-Nearest Neighbours Algorithm/KNN
Knowledge base
Knowledge Engineering
Knowledge graph
Knowledge Representation
Label space
Lagrange duality
Lagrange multiplier
Laplace smoothing
Laplacian correction
Latent Dirichlet Allocation/LDA 
Latent semantic analysis
Latent variable
Law of large number
Layer-wise Adaptive Rate Scaling/LARS
Lazy learning 
Leaky ReLU
Learner
Learning by analogy
Learning rate	
Learning Vector Quantization/LVQ	
Least squares regression tree
Leave-One-Out/LOO	
Lebesgue-integrable
Left eigenvector
Leibniz’s rule
Linear Discriminant Analysis/LDA	
Linear model
Linear Regression
Linear threshold units
Link function
Local conditional probability distribution
Local contrast normalization
Local curvature
Local Invariances
Local Markov property
Local minimum	
Log likelihood
Log odds/logit
Logistic Regression
Log-likelihood
Log-linear regression
Long-Short Term Memory/LSTM
Long-term dependency
Loopy belief propagation
Loss function
Low rank matrix approximation
Machine translation/MT
Macron-P	
Macron-R	
Main diagonal
Majority voting
Manifold assumption
Manifold learning
Manifold tangent classifier
Margin theory	
Marginal distribution
Marginal independence 	
Marginal probability distribution
Marginalization 	
Markov Chain
Markov Chain Monte Carlo/MCMC
Markov Random Field
Matrix inversion
Maximal clique
Maximum A Posteriori
Maximum Likelihood Estimation/MLE
Maximum margin	
Maximum weighted spanning tree
Max-Pooling	
Mean product of Student t-distribution
Mean squared error	
Mean-covariance restricted Boltzmann machine
Measure theory
Meta-learner
Metric learning
Micro-P	
Micro-R	
Mini-Batch SGD
Minimal  Description Length/MDL
Minimax game	
Misclassification cost
Mixture density network
Mixture of experts
Model predictive control (MPC)
Moment matching
Momentum	
Monte Carlo Estimate
Moore's Law
Moral graph	
Multi-class classification	
Multi-document summarization	
Multi-kernel learning
Multi-layer feedforward neural networks
Multilayer Perceptron/MLP
Multimodal learning
Multinomial distribution
Multiple Dimensional Scaling
Multiple linear regression
Multi-response Linear Regression/MLR	
Multivariate normal distribution
Mutual information
Naive bayes
Naive Bayes Classifier
Named entity recognition
Nash equilibrium
Nash reversion
Natural language generation/NLG
Natural language processing
Nearest-neighbor search
Negative class
Negative correlation
Negative definite
Negative Log Likelihood
Negative semidefinite
Neighbourhood Component Analysis/NCA
Neural Machine Translation
Neural Turing Machine
Neuromorphic Computing
Newton method
Conference on Neural Information Processing Systems/NIPS
No Free Lunch Theorem/NFL
Noise-contrastive estimation
Nominal attribute
Non-convex optimization
Nonlinear model
Non-linear oscillation
Non-metric distance
Non-negative matrix factorization
Non-ordinal attribute
Non-Saturating Game
Norm
Normalization
Nuclear norm
Numerical attribute
Numerical optimization
Objective function
Oblique decision tree
Occam's razor
Odds
Offline inference
Off-Policy
Offset vector
One shot learning
One-Dependent Estimator/ODE
Online inference
On-Policy
Ordinal attribute
Orthogonal matrix
Orthonormal
Outlier
Out-of-bag estimate	
Output layer
Output smearing
Overcomplete
Overfitting	
Oversampling
Paired t-test
Pairwise
Pairwise Markov property
Parallel tempering
Parameter
Parameter estimation
Parameter Server
Parameter tuning
Parse tree
Partial derivative
Particle Swarm Optimization/PSO
Part-of-speech tagging
Perceptron
Performance measure
Permutation invariant
Perplexity
Pictorial structure
Plug and Play Generative Network
Plurality voting
Polarity detection
Polynomial Basis Function
Polynomial kernel function
Pooling
Positive class
Positive definite matrix
Posterior inference
Posterior probability
Post-hoc test
Post-pruning
potential function
Power method
Precision
Prepruning
Principal component analysis/PCA
Principle of multiple explanations
Prior knowledge
Probability Graphical Model
Proximal Gradient Descent/PGD
Pruning
Pseudo-label
Quadratic Programming
Quantized Neural Network/QNN
Quantum computer
Quantum Computing
Quantum machine learning
Quasi Newton method
Quasi-concave
Radial Basis Function/RBF
Random Forest Algorithm
Random walk
Recall
Receiver Operating Characteristic/ROC
Rectified Linear Unit/ReLU
Recurrent Neural Network
Recursive neural network
Reference model
Regression
Regularization
Regularizer
Reinforcement learning/RL
Relative entropy
Reparametrization
Representation learning
Representer theorem
Reproducing Kernel Hilbert Space/RKHS
Re-sampling
Rescaling
Reservoir computing
Residual Blocks
Residual Mapping
Residual Network
Restricted Boltzmann Machine/RBM
Restricted Isometry Property/RIP
Reverse mode accumulation
Re-weighting
Ridge regression
Robustness
Root node
Rule Engine
Rule learning
Saddle point
Saddle-free Newton method
Saliency map
Sample space
Sampling
Score function
Second derivative
Second-order method
Self-contrastive estimation
Self-Driving
Self-Organizing Map/SOM
Semantic hashing
Semantic segmentation
Semantic similarity
Semi-Definite Programming
Semi-naive Bayes classifiers
Semi-restricted Boltzmann Machine
Semi-Supervised Learning
semi-Supervised Support Vector Machine
Sentiment analysis
Separating hyperplane
Shannon entropy
Shift invariance
Siamese Network
Sigmoid function
Similarity measure
Simulated annealing
Simultaneous localization and mapping/SLAM
Singular value
Singular Value Decomposition
Slack variables
Slowness principle
Smoothing
Smoothness prior
Soft margin
Soft margin maximization
Soft voting
Sparse activation
Sparse coding
Sparse connectivity
Sparse initialization
Sparse representation
Sparsity
Specialization
Spectral Clustering
Spectral radius
Speech Recognition
Spiking Neural Nets
Splitting variable
Squashing function
Stability-plasticity dilemma
Stacked Deconvolutional Network/SDN
Standard deviation
Static game
Stationary distribution
Stationary point
Statistical learning
Status feature function
Stochastic gradient descent
Stochastic Matrix
Stochastic maximum likelihood
Stochastic Neighbor Embedding
Stratified sampling
Structural  risk
Structural risk minimization/SRM
Structured variational inference
Subsampling
Subspace
Supervised learning
support vector expansion
Support Vector Machine/SVM
Surrogat loss
Surrogate function
Symbolic learning
Symbolism
Synset
Synthetic feature
Tangent plane
Tangent prop
T-Distribution Stochastic Neighbour Embedding/t-SNE
Tempered transition
Tensor
Tensor Processing Units/TPU
The least square method
Threshold
Threshold logic unit
Threshold-moving
Tiled convolution
Time delay neural network
Time Step
Tractable
Tokenization
Training error
Training instance
Transductive learning
Transfer learning
Treebank
Trial-by-error
Triangulate
Trigram
True negative
True positive
True Positive Rate/TPR
Turing Machine
Twice-learning
Two-dimensional array
Underestimation
Underfitting	
Undersampling	
Understandability	
Undirected graphical model
Unequal cost
Unit norm
Unit test 
Unit variance
Unitary matrix
Unit-step function
Univariate decision tree	
Unprojection
Unshared convolution
Unsupervised  learning
Unsupervised layer-wise training	
Upper Confidence Bounds
Upsampling	
Vanishing Gradient Problem
Variational derivative
Variational free energy
Variational inference
VC Theory
Version space
Virtual adversarial example
Viterbi algorithm
Von Neumann architecture
Weak learner
Weight
Weight decaying
Weight sharing
Weighted voting
Wasserstein GAN/WGAN
Within-class scatter matrix
Word embedding
Word sense disambiguation
Zero mean
Zero-data learning
Zero-shot learning