blanketglossary

Self-supervised learning

Definition

Self-supervised learning (SSL) is a paradigm in machine learning where a model is trained on a task using the data itself to generate supervisory signals, rather than relying on externally-provided labels. In the context of neural networks, self-supervised learning aims to leverage inherent structures or relationships within the input data to create meaningful training signals. SSL tasks are designed so that solving them requires capturing essential features or relationships in the data. The input data is typically augmented or transformed in a way that creates pairs of related samples, where one sample serves as the input, and the other is used to formulate the supervisory signal. This augmentation can involve introducing noise, cropping, rotation, or other transformations. Self-supervised learning more closely imitates the way humans learn to classify objects.

Related concepts

15.aiAAAI Conference on Artificial IntelligenceAI agentAI alignmentAI anthropomorphismAI boomAI bubbleAI data centerAI effectAI literacyAI nationalismAI safetyAI slopAI takeoverAI veganismAI winterAction selectionActivation functionActive learning (machine learning)Adaptive learningAdobe FireflyAdversarial machine learningAgent2AgentAidan GomezAlan TuringAlexNetAlex Graves (computer scientist)Alex KrizhevskyAllen NewellAlphaFoldAlphaGoAlphaZeroAndrej KarpathyAndrew NgAnomaly detectionApplications of artificial intelligenceApprenticeship learningArtificial Intelligence ActArtificial Intelligence Cold WarArtificial general intelligenceArtificial human companionArtificial intelligenceArtificial intelligence and electionsArtificial intelligence arms raceArtificial intelligence in architectureArtificial intelligence in educationArtificial intelligence in fictionArtificial intelligence in healthcareArtificial intelligence in mental healthArtificial intelligence in video gamesArtificial intelligence visual artArtificial neural networkArtificial superintelligenceAshish VaswaniAssociation rule learningAttention (machine learning)Audio signal processingAurora (text-to-image model)AutoGPTAutoencoderAutoencodersAutomated machine learningAutomated reasoningAutomated theorem provingAutoregressive modelBERT (language model)BIRCHBLOOM (language model)BackpropagationBatch learningBatch normalizationBayesian networkBernard WidrowBias–variance tradeoffBinary classificationBoltzmann machineBoosting (machine learning)Bootstrap aggregatingCURE algorithmCanonical correlationChatbot psychosisChinchilla (language model)Christopher D. ManningClaude (language model)Claude ShannonCliff ShawCluster analysisCoefficient of determinationCompetition in artificial intelligenceComputational learning theoryComputer visionConcept driftConditional random fieldConference on Neural Information Processing SystemsConfusion matrixConjugate gradient methodContrastive Language-Image Pre-trainingConvolutionConvolutional neural networkConvolutional neural networksCosine similarityCrowdsourcingCurriculum learningDALL-EDBRXDBSCANDaniel Kokotajlo (researcher)Data augmentationData cleaningData miningDavid Silver (computer scientist)Decision tree learningDeepDreamDeepSeek (chatbot)Deep learningDeep learning speech synthesisDemis HassabisDensity estimationDifferentiable neural computerDiffusion modelDiffusion processDimensionality reductionDomain knowledgeDouble descentDream Machine (text-to-video model)ECML PKDDEcho state networkElectrochemical RAMElevenLabsEmpirical risk minimizationEnsemble learningEnvironmental impact of artificial intelligenceEthics of artificial intelligenceExpectation–maximization algorithmExplainable artificial intelligenceFacebookFacial recognition systemFactor analysisFeature (machine learning)Feature engineeringFeature learningFeedforward neural networkFei-Fei LiFlux (text-to-image model)Frank RosenblattFrançois CholletFuzzy clusteringGPT-3GPT ImageGated recurrent unitGating mechanismGemini (chatbot)Gemini (language model)Gemma (language model)Generative AIGenerative adversarial networkGenerative engine optimizationGenerative modelGenerative pre-trained transformerGenie (world model)Geoffrey HintonGloVeGlossary of artificial intelligenceGoogleGradient descentGrammar inductionGraph neural networkGraphical modelGrok (chatbot)Hallucination (artificial intelligence)Handwriting recognitionHerbert A. SimonHidden Markov modelHierarchical clusteringHighway networkHistory of artificial intelligenceHuawei PanGuHuman-in-the-loopHuman image synthesisHumanity's Last ExamHyperparameter (machine learning)IBM GraniteIBM WatsonIBM WatsonxIan GoodfellowIdeogram (text-to-image model)Ilya SutskeverImageNetImagen (text-to-image model)Imitation learningIndependent component analysisIntelligent agentInternational Conference on Learning RepresentationsInternational Conference on Machine LearningInternational Joint Conference on Artificial IntelligenceIsolation forestJames GoodnightJan LeikeJohn HopfieldJohn McCarthy (computer scientist)John SchulmanJohn von NeumannJoseph WeizenbaumJournal of Machine Learning ResearchJürgen SchmidhuberK-means clusteringK-nearest neighbors algorithmKernel machinesKling AIKunihiko FukushimaLaMDALabeled dataLanguage modelLarge language modelLatent diffusion modelLatent spaceLeNetLearning curve (machine learning)Learning to rankLethal autonomous weaponLinear discriminant analysisLinear regressionList of artificial intelligence companiesList of artificial intelligence projectsList of datasets for machine-learning researchList of datasets in computer vision and image processingLlama (language model)Local outlier factorLogistic regressionLong short-term memoryLoss functionLoss functions for classificationLotfi A. ZadehMachine Learning (journal)Machine learningMamba (deep learning architecture)Marvin MinskyMean shiftMean squared errorMechanistic interpretabilityMemtransistorMeta-learning (computer science)MidjourneyMiniMax (company)Model Context ProtocolMuZeroMulti-agent reinforcement learningMultilayer perceptronMultimodal learningMusic and artificial intelligenceMustafa SuleymanNaive Bayes classifierNathaniel Rochester (computer scientist)Natural language processingNeural Turing machineNeural fieldNeural machine translationNeural network (machine learning)Neural radiance fieldNeuro-symbolic AINeuromorphic engineeringNoam ShazeerNon-negative matrix factorizationNormalization (machine learning)OPTICS algorithmOasis (Minecraft clone)Occam learningOliver SelfridgeOnline machine learningOntology learningOpenAIOpenAI FiveOptical character recognitionOriol VinyalsOutline of machine learningOverfittingPaLMParameterPaul WerbosPerceptronPhysics-informed neural networksPolicy gradient methodPolysemyPrecautionary principlePrincipal component analysisProbably approximately correct learningProject DebaterPrompt engineeringProper generalized decompositionQ-learningQuantum machine learningQuasi-Newton methodQuoc V. LeQwenRandom forestRandom sample consensusReasoning modelReceiver operating characteristicRecraftRectifier (neural networks)Recurrent neural networkRecursive self-improvementReflection (artificial intelligence)Regression analysisRegularization (mathematics)Regulation of artificial intelligenceRegulation of artificial intelligence in the United StatesReinforcement learningReinforcement learning from human feedbackRelevance vector machineReservoir computingResidual neural networkRestricted Boltzmann machineRetrieval-augmented generationRiffusionRobot controlRule-based machine learningRunway (company)Seedance 2.0Self-driving carSelf-organizing mapSelf-play (reinforcement learning technique)Semantic analysis (machine learning)Semi-supervised learningSeppo LinnainmaaSeq2seqSeymour PapertShun'ichi AmariSigmoid functionSoftmax functionSora (text-to-video model)Sparse dictionary learningSpeech recognitionSpiking neural networkStable DiffusionState–action–reward–state–actionStatistical classificationStatistical learning theoryStephen GrossbergStochastic gradient descentStructured predictionSuno (platform)Supervised learningSupport vector machineSymbolic artificial intelligenceT-distributed stochastic neighbor embeddingT5 (language model)Takeo KanadeTemporal difference learningText-to-image modelText-to-video modelTimeline of artificial intelligenceTopological deep learningTraining, validation, and test data setsTransfer learningTransformer (deep learning)Transformer (deep learning architecture)U-NetUdioUncanny valleyUnsupervised learningVapnik–Chervonenkis theoryVariational autoencoderVeo (text-to-video model)Vibe codingVirtual politicianVision transformerWalter PittsWarren Sturgis McCullochWaveNetWeak artificial intelligenceWeight initializationWhisper (speech recognition system)Word2vecWord embeddingWord sense disambiguationWorkplace impact of artificial intelligenceWorld model (artificial intelligence)Xiaomi MiMoYann LeCunYarowsky algorithmYoshua Bengio

17 concepts already in your glossary