Machine learning terms

See also: Terms and Machine learning Fundamentals** Natural Language Processing - Language Evaluation**** Fairness**** Decision Forests**** Reinforcement Learning**** Computer Vision** - Image Models** Sequence Models**** Clustering**** Recommendation Systems**** TensorFlow**** Google Cloud**

All ML Terms

See also: Machine learning terms/All

• A/B testing

• accuracy

• action

• activation function

• active learning

• AdaGrad

• agent

• agglomerative clustering

• anomaly detection

• AR

• area under the PR curve

• area under the ROC curve

• artificial general intelligence

• artificial intelligence

• attention

• attribute

• attribute sampling

• AUC (Area under the ROC curve)

• augmented reality

• automation bias

• average precision

• axis-aligned condition

• backpropagation

• bagging

• bag of words

• baseline

• batch

• batch normalization

• batch size

• Bayesian neural network

• Bayesian optimization

• Bellman equation

• BERT (Bidirectional Encoder Representations from Transformers)

• bias (ethics/fairness)

• bias (math) or bias term

• bigram

• bidirectional

• bidirectional language model

• binary classification

• binary condition

• binning

• BLEU (Bilingual Evaluation Understudy)

• boosting

• bounding box

• broadcasting

• bucketing

• calibration layer

• candidate generation

• candidate sampling

• categorical data

• causal language model

• centroid

• centroid-based clustering

• checkpoint

• class

• classification model

• classification threshold

• class-imbalanced dataset

• clipping

• Cloud TPU

• clustering

• co-adaptation

• collaborative filtering

• condition

• confirmation bias

• confusion matrix

• continuous feature

• convenience sampling

• convergence

• convex function

• convex optimization

• convex set

• convolution

• convolutional filter

• convolutional layer

• convolutional neural network

• convolutional operation

• cost

• co-training

• counterfactual fairness

• coverage bias

• crash blossom

• critic

• cross-entropy

• cross-validation

• data analysis

• data augmentation

• DataFrame

• data parallelism

• data set or dataset

• Dataset API (tf.data)

• decision boundary

• decision forest

• decision threshold

• decision tree

• deep model

• decoder

• deep neural network

• Deep Q-Network (DQN)

• demographic parity

• denoising

• dense feature

• dense layer

• depth

• depthwise separable convolutional neural network (sepCNN)

• derived label

• device

• dimension reduction

• dimensions

• discrete feature

• discriminative model

• discriminator

• disparate impact

• disparate treatment

• divisive clustering

• downsampling

• DQN

• dropout regularization

• dynamic

• dynamic model

• eager execution

• early stopping

• earth mover's distance (EMD)

• embedding layer

• embedding space

• embedding vector

• empirical risk minimization (ERM)

• encoder

• ensemble

• entropy

• environment

• episode

• epoch

• epsilon greedy policy

• equality of opportunity

• equalized odds

• Estimator

• example

• experience replay

• experimenter's bias

• exploding gradient problem

• fairness constraint

• fairness metric

• false negative (FN)

• false negative rate

• false positive (FP)

• false positive rate (FPR)

• feature

• feature cross

• feature engineering

• feature extraction

• feature importances

• feature set

• feature spec

• feature vector

• federated learning

• feedback loop

• feedforward neural network (FFN)

• few-shot learning

• fine tuning

• forget gate

• full softmax

• fully connected layer

• GAN

• generalization

• generalization curve

• generalized linear model

• generative adversarial network (GAN)

• generative model

• generator

• GPT (Generative Pre-trained Transformer)

• gini impurity

• gradient

• gradient boosting

• gradient boosted (decision) trees (GBT)

• gradient clipping

• gradient descent

• graph

• graph execution

• greedy policy

• ground truth

• group attribution bias

• hallucination

• hashing

• heuristic

• hidden layer

• hierarchical clustering

• hinge loss

• holdout data

• hyperparameter

• hyperplane

• i.i.d.

• image recognition

• imbalanced dataset

• implicit bias

• incompatibility of fairness metrics

• independently and identically distributed (i.i.d)

• individual fairness

• inference

• inference path

• information gain

• in-group bias

• input layer

• in-set condition

• instance

• interpretability

• inter-rater agreement

• intersection over union (IoU)

• IoU

• item matrix

• items

• iteration

• Keras

• keypoints

• Kernel Support Vector Machines (KSVMs)

• k-means

• k-median

• L0 regularization

• L1 loss

• L1 regularization

• L2 loss

• L2 regularization

• label

• labeled example

• LaMDA (Language Model for Dialogue Applications)

• lambda

• landmarks

• language model

• large language model

• layer

• Layers API (tf.layers)

• leaf

• learning rate

• least squares regression

• linear model

• linear

• linear regression

• logistic regression

• logits

• Log Loss

• log-odds

• Long Short-Term Memory (LSTM)

• loss

• loss curve

• loss function

• loss surface

• LSTM

• machine learning

• majority class

• Markov decision process (MDP)

• Markov property

• masked language model

• matplotlib

• matrix factorization

• Mean Absolute Error (MAE)

• Mean Squared Error (MSE)

• metric

• meta-learning

• Metrics API (tf.metrics)

• mini-batch

• mini-batch stochastic gradient descent

• minimax loss

• minority class

• ML

• MNIST

• modality

• model

• model capacity

• model parallelism

• model training

• Momentum

• multi-class classification

• multi-class logistic regression

• multi-head self-attention

• multimodal model

• multinomial classification

• multinomial regression

• NaN trap

• natural language understanding

• negative class

• neural network

• neuron

• N-gram

• NLU

• node (neural network)

• node (TensorFlow graph)

• node (decision tree)

• noise

• non-binary condition

• nonlinear

• non-response bias

• nonstationarity

• normalization

• novelty detection

• numerical data

• NumPy

• objective

• objective function

• oblique condition

• offline

• offline inference

• one-hot encoding

• one-shot learning

• one-vs.-all

• online

• online inference

• operation (op)

• out-of-bag evaluation (OOB evaluation)

• optimizer

• out-group homogeneity bias

• outlier detection

• outliers

• output layer

• overfitting

• oversampling

• pandas

• parameter

• Parameter Server (PS)

• parameter update

• partial derivative

• participation bias

• partitioning strategy

• perceptron

• performance

• permutation variable importances

• perplexity

• pipeline

• pipelining

• policy

• pooling

• positive class

• post-processing

• PR AUC (area under the PR curve)

• precision

• precision-recall curve

• prediction

• prediction bias

• predictive parity

• predictive rate parity

• preprocessing

• pre-trained model

• prior belief

• probabilistic regression model

• proxy (sensitive attributes)

• proxy labels

• Q-function

• Q-learning

• quantile

• quantile bucketing

• quantization

• queue

• random forest

• random policy

• ranking

• rank (ordinality)

• rank (Tensor)

• rater

• recall

• recommendation system

• Rectified Linear Unit (ReLU)

• recurrent neural network

• regression model

• regularization

• regularization rate

• reinforcement learning (RL)

• ReLU

• replay buffer

• reporting bias

• representation

• re-ranking

• return

• reward

• ridge regularization

• RNN

• ROC (receiver operating characteristic) Curve

• root

• root directory

• Root Mean Squared Error (RMSE)

• rotational invariance

• sampling bias

• sampling with replacement

• SavedModel

• Saver

• scalar

• scaling

• scikit-learn

• scoring

• selection bias

• self-attention (also called self-attention layer)

• self-supervised learning

• self-training

• semi-supervised learning

• sensitive attribute

• sentiment analysis

• sequence model

• sequence-to-sequence task

• serving

• shape (Tensor)

• shrinkage

• sigmoid function

• similarity measure

• size invariance

• sketching

• softmax

• sparse feature

• sparse representation

• sparse vector

• sparsity

• spatial pooling

• split

• splitter

• squared hinge loss

• squared loss

• stability

• staged training

• state

• state-action value function

• static

• static inference

• stationarity

• step

• step size

• stochastic gradient descent (SGD)

• stride

• structural risk minimization (SRM)

• subsampling

• summary

• supervised machine learning

• synthetic feature

• tabular Q-learning

• target

• target network

• temporal data

• Tensor

• TensorBoard

• TensorFlow

• TensorFlow Playground

• TensorFlow Serving

• Tensor Processing Unit (TPU)

• Tensor rank

• Tensor shape

• Tensor size

• termination condition

• test

• test loss

• test set

• tf.Example

• tf.keras

• threshold (for decision trees)

• time series analysis

• timestep

• token

• tower

• TPU

• TPU chip

• TPU device

• TPU master

• TPU node

• TPU Pod

• TPU resource

• TPU slice

• TPU type

• TPU worker

• training

• training loss

• training-serving skew

• training set

• trajectory

• transfer learning

• Transformer

• translational invariance

• trigram

• true negative (TN)

• true positive (TP)

• true positive rate (TPR)

• unawareness (to a sensitive attribute)

• underfitting

• undersampling

• unidirectional

• unidirectional language model

• unlabeled example

• unsupervised machine learning

• uplift modeling

• upweighting

• user matrix

• validation

• validation loss

• validation set

• vanishing gradient problem

• variable importances

• Wasserstein loss

• weight

• Weighted Alternating Least Squares (WALS)

• weighted sum

• wide model

• width

• wisdom of the crowd

• word embedding

• Z-score normalization