AI Glossary
0-9
A
A/B TestingAccountabilityAccuracyAcoustic ModelingActivation FunctionsActive LearningActor-Critic MethodsActuatorsAdaDeltaAdaGradAdam OptimizerAdjusted R-SquaredAdversarial AttacksAffordance LearningAgent-Based ModelingAgentic AI / Autonomous AgentsAgentic AI FrameworksAgglomerative ClusteringAI AcceleratorsAI Act (EU)AI AgentsAI AlignmentAI and BiasAI and SustainabilityAI APIsAI Art GenerationAI AssistantsAI AuditAI AuditingAI Bill of Rights (US Blueprint)AI ContainmentAI DemocratizationAI Ethics BoardsAI Ethics GuidelinesAI Feature StoreAI for Climate ChangeAI Generated ContentAI Governance FrameworksAI GuardrailsAI HallucinationsAI in Healthcare EthicsAI in WarfareAI LegislationAI LiteracyAI MarketplacesAI Model GovernanceAI Model HubAI Model RegistryAI Model WeightsAI Music GenerationAI OrchestrationAI PolicyAI RegulationsAI SafetyAI SecurityAI SingularityAI Transparency ReportAI WatermarkingAI WinterAI Workflow AutomationAI-as-a-ServiceAlan TuringAlgorithmic AccountabilityAlgorithmic Bias MitigationAlgorithmic DiscriminationAlgorithmic TransparencyAndrew NgAnomaly DetectionAnomaly Detection in SecurityAnthropicApache KafkaAPI DevelopmentAPI EndpointsApriori AlgorithmArtificial General Intelligence (AGI)Artificial Neural NetworksArtificial SuperintelligenceASICsAssociation Rule LearningAsynchronous Advantage Actor-CriticAttention MechanismsAUCAudio ClassificationAudio Signal ProcessingAugmented RealityAuthenticationAuthorizationAutoencodersAutomated ReasoningAutomatic Speech Recognition (ASR)AutomationAutoMLAutonomous NavigationAutoregressive ModelsWhat is AGI / Artificial General IntelligenceWhat is AlgorithmWhat is Artificial Intelligence (AI)What is AttentionWhat is Autoencoder
B
Bag-of-Words ModelBaggingBatch SizeBayesian InferenceBayesian NetworksBayesian OptimizationBias in AIBias-Variance TradeoffBig DataBig Data TechnologiesBiometric SecurityBLEU ScoreBlockchain in AIBox PlotByte-Pair Encoding (BPE)What is BackpropagationWhat is Batch NormalizationWhat is BERTWhat is BiasWhat is Boosting
C
CaffeCalculusCalibrationCalifornia Consumer Privacy Act (CCPA)Canary DeploymentCapsule NetworksCarbon Footprint of AICase-Based ReasoningCatastrophic ForgettingCentral Limit TheoremChain-of-ThoughtChinese Room ArgumentClass ImbalanceClassificationCloud AI PlatformsCloud ComputingClustering AlgorithmsCode Generation ModelsCognitive ArchitecturesCognitive ComputingCohereColab NotebooksCollaborative FilteringColor SpacesComplex AnalysisComplianceCompliance Standards (ISO IEEE)Computational ComplexityComputational Fluid DynamicsComputational Theory of MindCompute-Optimal ModelsConcept DriftConceptual GraphsConditional ProbabilityConfusion MatrixConsciousness in AIConsistency ModelsConstitutional AIConstraint Satisfaction ProblemsContainerizationContent-Based FilteringContext WindowContinual LearningContinuous Integration/Continuous Deployment (CI/CD)Control SystemsConversational AIConvolutional Neural NetworksCOPPACoreference ResolutionCorrelationCorrelation MatrixCost-Sensitive LearningCross-Entropy LossCurriculum LearningCyber Threat IntelligenceCybersecurity RegulationsWhat is ChatbotWhat is Classifier / ClassificationWhat is ClusteringWhat is CNN / Convolutional Neural NetworkWhat is Cross-Validation
D
DALL·EData AnnotationData CatalogData CentersData CleaningData DriftData GovernanceData IngestionData IntegrationData LabelingData LakeData LakesData LeakageData LineageData MiningData PipelineData PoisoningData PreprocessingData PrivacyData ProtectionData Protection LawsData QualityData SecurityData SovereigntyData TransformationData VersioningData VisualizationData Visualization TechniquesData WarehousingDatabases for AIDavies-Bouldin IndexDBSCANDecision Boundary VisualizationDecision TreesDeep Belief NetworksDeep Q-NetworksDeep Reinforcement LearningDeepfakesDeepMindDemis HassabisDependency ParsingDepth EstimationDescriptive StatisticsDialogue SystemsDifferential EquationsDifferential EvolutionDifferential PrivacyDiffusion ModelsDigital DivideDigital ProvenanceDigital TwinsDimensionality ReductionDirect Preference Optimization (DPO)Discourse AnalysisDiscrete Event SimulationDiscrete MathematicsDisinformationDistributed ComputingDistributed File SystemsDistributed TrainingDockerDronesDropoutDropout RegularizationDynamical SystemsWhat is Data AugmentationWhat is Deep LearningWhat is DeepfakeWhat is Deterministic ModelWhat is Discriminative Model
E
Early StoppingEdge AIEdge ComputingEdge DetectionEigenvalues and EigenvectorsElon MuskEmbedding SizeEmbeddingsEmbodied AIEmergent AbilitiesEmotion RecognitionEnsemble MethodsEpisodic MemoryEthical AIEthical AI GuidelinesEthical AuditingEthical Decision-MakingEthical DilemmasEthical FrameworksEthics of AIETL ProcessesEvolutionary AlgorithmsExistential RiskExpectation-MaximizationExpectation-Maximization AlgorithmExpected Calibration ErrorExpert SystemsExplainabilityExploration vs. ExploitationExploratory Data AnalysisExport ControlsWhat is EmbeddingWhat is EncoderWhat is Ensemble LearningWhat is EpochWhat is Explainable AI (XAI)
F
F1 ScoreFacial RecognitionFairnessFastAIFeature EngineeringFeature ImportanceFeature SelectionFeature StoreFeature StoresFederated LearningFei-Fei LiFew-Shot LearningFinite Element AnalysisFirst-Order LogicFlow MatchingForce ControlFoundation Model EconomyFoundation ModelsFourier TransformFPGAsFrame LanguagesFunctional AnalysisWhat is Feature ExtractionWhat is Fine-tuningWhat is Forward PropagationWhat is Foundation ModelWhat is Fusion / Multimodal Fusion
G
Game Playing AIGame TheoryGame Theory SimulationsGated Recurrent UnitsGaussian Mixture ModelsGeneral Data Protection Regulation (GDPR)Generative Adversarial NetworksGenerative ModelsGenetic AlgorithmsGensimGeoffrey HintonGlobal CooperationGPT ModelsGrad-CAMGradient Boosting MachinesGradient ClippingGraph Neural NetworksGraph TheoryGraphics Processing Units (GPUs)Grid SearchWhat is GAN / Generative Adversarial NetworkWhat is Generative AIWhat is Gradient DescentWhat is Graph Neural Network (GNN)What is Grounding
H
HadoopHeatmapHelpHeuristic AlgorithmsHidden Markov ModelsHierarchical Reinforcement LearningHigh-Performance ComputingHIPAAHistogramHOGHPC ClustersHugging FaceHugging Face TransformersHuman RightsHuman-in-the-LoopHuman-Robot InteractionHyperparameter OptimizationHyperparameter TuningWhat is HallucinationWhat is HeuristicWhat is Hidden LayerWhat is Hierarchical ModelWhat is Hyperparameter
I
Ilya SutskeverImage CaptioningImage ClassificationImage RecognitionImage SegmentationImpact on EmploymentIn-Context LearningIndustrial RobotsInferenceInference EnginesInference OptimizationInferential StatisticsInformation TheoryInformed ConsentInfrastructure as CodeInstance SegmentationIntellectual Property RightsIntelligent AgentsIntrusion Detection SystemsInverse Reinforcement LearningWhat is Imbalanced DataWhat is Instance / SampleWhat is Instruction tuningWhat is Intelligence Amplification / AugmentationWhat is Interpretability
J
K
L
L1 RegularizationL2 RegularizationLabel SmoothingLanguage ModelingLanguage ModelsLaplace TransformLarge Language Models (LLMs)Large Multimodal ModelsLatent Dirichlet AllocationLatent SpaceLaw of Large NumbersLayer NormalizationLearning CurveLearning Rate DecayLearning Rate SchedulingLemmatizationLIMELinear AlgebraLinear RegressionLog LossLogic ProgrammingLogistic RegressionLong Short-Term Memory NetworksLong-Context ModelsLoRA (Low-Rank Adaptation)What is Large Language Model (LLM)What is Latent VariableWhat is Learning RateWhat is Loss FunctionWhat is LSTM / Long Short-Term Memory
M
Machine ConsciousnessMachine TranslationMarkov Chain ModelsMarkov Chain Monte CarloMarkov Decision ProcessesMarkov ModelsMarvin MinskyMasked Language ModelsMaster Data ManagementMatplotlibMatrix DecompositionMCPMean Absolute ErrorMean Squared ErrorMechanistic InterpretabilityMel-Frequency Cepstral Coefficients (MFCCs)Metadata ManagementMicroservicesMidjourneyMind UploadingMini ToolMini-Batch Gradient DescentMixture of Experts (MoE)MLOpsMobile RobotsModel CardsModel CompressionModel DeploymentModel DriftModel Explainability ToolsModel MonitoringModel ServingModel StealingMomentum OptimizationMonitoring and LoggingMonte Carlo MethodsMonte Carlo SimulationsMoral MachinesMotion DetectionMotion PlanningMulti-Armed Bandit ProblemMultimodal AIMusic Information RetrievalMXNetWhat is Machine Learning (ML)What is Meta-learningWhat is ModelWhat is Multi-head AttentionWhat is Multimodal / Multimodality
N
n-GramsNaive Bayes AlgorithmNaive Bayes ClassifierNamed Entity RecognitionNatural Language Generation (NLG)Natural Language ProcessingNatural Language Processing (NLP)Natural Language UnderstandingNesterov Accelerated GradientNetwork SimulationsNeural Architecture SearchNeural NetworksNeural Processing Unit (NPU)Neuromorphic ComputingNick BostromNLTKNoise ReductionNoSQL DatabasesNumPyNVIDIA CUDAWhat is Neural NetworkWhat is NLP / Natural Language ProcessingWhat is NLU / Natural Language UnderstandingWhat is NormalizationWhat is Novelty Detection / Anomaly Detection
O
P
PandasParallel ComputingParameter CountParameter-Efficient Fine-Tuning (PEFT)Part-of-Speech TaggingPartial Dependence PlotsPath PlanningPattern RecognitionPeople also viewedPerception in AIPerceptronPerplexityPeter NorvigPhilosophy of MindPhoneticsPipelinesPlanning and SchedulingPlotlyPolicy GradientsPolicy OptimizationPose EstimationPositional EncodingPragmaticsPrecisionPredictive ModelingPredictive ProbabilityPreference TuningPrincipal Component AnalysisPrivacyPrivacy-Preserving Machine LearningProbability Density FunctionsProbability TheoryProblem SolvingProcess ModelingProcess-Based SupervisionPrompt ChainingPrompt EngineeringPrompt InjectionPrompt MarketplacePrompt TemplatesPropositional LogicProximal Policy OptimizationPruningPyTorchWhat is ParameterWhat is Policy / Reinforcement Learning PolicyWhat is PoolingWhat is PretrainingWhat is Prompt
Q
R
R-SquaredRandom ForestsRandom SearchRay KurzweilReal AnalysisReasoning EnginesRecallRecommender SystemsRecurrent Neural NetworksRed TeamingRegressionRegression AnalysisRegulatory ComplianceReinforcement Learning from Human FeedbackReinforcement Learning in RoboticsReproducibilityResponsible AIRetrieval-Augmented GenerationReward FunctionRMSpropRobot KinematicsRobot VisionRobotic ManipulationRobotic Operating System (ROS)Robotics TransformersRobustness in AI ModelsROC CurveRodney BrooksRoot Mean Squared ErrorRule-Based SystemsWhat is RegularizationWhat is Reinforcement Learning (RL)What is Representation LearningWhat is Retrieval Augmented Generation (RAG)What is RNN / Recurrent Neural Network
S
Saliency MapsSARSA AlgorithmScalable OversightScaling LawsScatter PlotScikit-LearnSciPySeabornSearch AlgorithmsSecure HardwareSecure Multi-Party ComputationSecure ProtocolsSelf-AttentionSelf-Driving CarsSemantic NetworksSemantic ParsingSemantic Role LabelingSemantic SegmentationSemantic WebSemi-Supervised LearningSensorsSentencePieceSentiment AnalysisSequence LabelingServerless ComputingServerless GPUsSet TheorySHAP ValuesSiamese NetworksSIFTSilhouette ScoreSimulated AnnealingSimulation HypothesisSimulation-to-Real Transfer (Sim2Real)Simultaneous Localization and Mapping (SLAM)SMOTESocial Acceptance of AISocial SimulationSOTA (State of the Art)spaCySparkSpeaker DiarizationSpectrogram AnalysisSpeech EnhancementSpeech RecognitionSpeech SynthesisSpiking Neural NetworksSQLStable DiffusionStackingState-Action PairsStatistical AnalysisStatistical DistributionsStatisticsStemmingStochastic Gradient DescentStochastic ModelingStochastic ProcessesStop WordsStream ProcessingStrong AIStrong vs. Weak AIStuart RussellStyle TransferSubword TokenizationSupport Vector MachinesSURFSurveillanceSwarm IntelligenceSymbolic AISynthetic Data GenerationSynthetic MediaSystem DynamicsSystem PromptWhat is SamplingWhat is Self-Supervised LearningWhat is Sequence ModelingWhat is SoftmaxWhat is Supervised Learning
T
t-SNETeacher ForcingTechnological SingularityTeleoperationTemperatureTemporal Difference LearningTensor Processing Units (TPUs)TensorFlowTesting and ValidationText SummarizationText-to-Audio GenerationText-to-Image GenerationText-to-Speech (TTS)Text-to-Video GenerationTF-IDFTheanoTime Series AnalysisTimnit GebruTinyMLToken LimitTokenizationTokensTool Use (LLMs)Topic ModelingTopologyTransformer ModelsTransformer NetworksTransparencyTransparency RequirementsTrust Region Policy OptimizationTrustworthy AITruthfulness (in LLMs)Turing TestWhat is TokenizerWhat is Training DataWhat is Transfer LearningWhat is TransformerWhat is Tuning / Hyperparameter Tuning
U
V
Validation CurveValue FunctionVector DatabaseVersion Control for ModelsVibe code an AI ToolVideo Generation ModelsVirtual Reality SimulationsVoice BiometricsVoice CloningVoice ConversionWhat is Validation SetWhat is Vanishing / Exploding GradientWhat is Variational Autoencoder (VAE)What is Vector EmbeddingWhat is Vision Transformer (ViT)
W
X
Y
Mixture of Experts (MoE)
Mixture of Experts (MoE) is a machine learning architecture that dynamically selects a subset of models to make predictions based on the input data. This approach allows for efficient computation and can improve performance by leveraging specialized models for different tasks. In MoE, multiple 'expert' models are trained, but only a few are activated during inference, which helps manage resources and enhance scalability. Common use cases include natural language processing, computer vision, and other areas where diverse expertise can lead to better predictions and insights.