Skip to content
MLE Path
Interview prepPractice from reported loops, plans and peer mocks.ConceptsEvery ML concept, with papers, engineering blogs and docs.JobsML roles, employers and resume scan.PricingFree account and Career+.
Account
Concepts

Explore

Concept mapTopic mapsLearning pathsModules

Practice

AssessmentsKnowledge testsYour learning
← All concepts
Browse areas
Finding an ML job208
Finding an ML job
Career and interviewing161
Career and interviewing
Career document22
Career documentApplication trackerBrag documentCareer level frameworkCover letterCurriculum vitaeEmployment negotiation planEmployment offerEngineering evidence portfolioGrowth planInterview feedback recordInterview rubricJob descriptionJob requirement and evidence matrixJob-search constraintsLearning agendaLearning planPromotion documentReferral requestResumeRole transition planTarget role profile
Job interviewing67
Job interviewing
Interview loop3
Interview loopAmazon interview loopMeta full loop interview
Interview preparation2
Interview preparationInterview practice
Interview round23
Interview round
Behavioral interview4
Behavioral interviewAmazon Leadership Principles interviewGoogle behavioral and leadership interview contextPeople management interview
Coding interview5
Coding interviewAI-assisted coding interviewAlgorithmic coding interviewLow-level design interviewML coding interview
System design interview7
System design interview
Distributed system design interview2
Distributed system design interviewMeta Systems Design interview
AI application design interviewMeta Product Architectural Design interviewML platform design interviewML system design interview
Amazon SDE online assessmentAmazon writing exerciseHiring manager interviewProject deep-dive interview roundRecruiter screenTeam matching interview
Most asked in interviews, by round45
Coding rounds19
Most asked in coding rounds14
NumPyAlgorithmsData structuresGraph theoryCross-entropy lossInvariantNumerical stabilityBackpropagationPython programmingBinary searchDepth-first searchHash tableProbability theorySorting
Algorithmic coding interviewML coding interviewAI-assisted coding interviewLow-level design interviewCoding interview
ML fundamentals, math and evaluation rounds0
Most asked in ML fundamentals, math and evaluation rounds0
System design rounds16
Most asked in system design rounds11
Information retrievalRetrieval-augmented generationLearning to rankAccess controlAgent systemsApproximate nearest-neighbor searchDecision-threshold selectionDistributed systemsEvidence groundingIdempotencyTwo-tower retrieval model
ML system design interviewAI application design interviewDistributed system design interviewML platform design interviewSystem design interview
Performance and infrastructure rounds0
Most asked in performance and infrastructure rounds0
Project, behavioral and leadership rounds10
Most asked in project, behavioral and leadership rounds6
Present career evidenceNavigate an interviewCommunicate uncertaintyManage stakeholder relationshipsEvaluate a prospective teamPast-behavior interview question
Project deep-dive interview roundExplain a project in depthBehavioral interviewPeople management interview
Calibrate interview evidenceConduct a behavioral interviewConduct a technical interviewExplain a project in depthNavigate an interviewPast-behavior interview questionSituational interview questionStructured employment interviewWork-sample assessment
Job search27
Job search
Total compensation8
Total compensation
Equity compensation5
Equity compensationEmployee stock optionEquity liquidity riskRestricted stock unitVesting schedule
Base salarySigning bonus
Job-search funnel2
Job-search funnelApplication-stage conversion rate
Applicant tracking systemApplication trackingApplication-stage conversion rateEmployee referralEmployment-offer comparisonEvaluate a prospective teamInterpret job requirementsJob applicationJob-search funnelOffer acceptancePreferred job qualificationRequired job qualificationResume evidence reviewResume keyword stuffingResume readability reviewResume-to-job evidence alignmentRole targetingSalary and employment-offer negotiation
ML job roles31
ML job roles
Data and platform engineering job family4
Data and platform engineering job familyAnalytics engineerData engineerML platform engineer
Data science and experimentation job family4
Data science and experimentation job familyData scientistExperimentation and causal inference scientistForecasting scientist
ML engineering job family8
ML engineering job familyComputer vision engineerInference and serving engineerLLM application engineerMachine learning engineerNLP engineerRecommendation and ranking engineerResearch engineer
Reliability engineering job family2
Reliability engineering job familyML reliability engineer
Research science job family3
Research science job familyApplied scientistML research scientist
Robotics engineering job family2
Robotics engineering job familyRobotics ML engineer
Security, privacy and governance job family3
Security, privacy and governance job familyML security engineerPrivacy engineer for ML
Technical leadership and management job family4
Technical leadership and management job familyML engineering managerML product leadStaff-level ML engineer
Present career evidence3
Present career evidenceCompose application materialsPrepare a promotion document
Professional network2
Professional networkWeak professional tie
Professional networking2
Professional networkingInformational interview
Promotion review2
Promotion reviewPromotion calibration
Diagnose learning gapsEngineer onboardingLearn an unfamiliar domainPlan a role transitionPractice with feedbackRole transition planning
Communication and collaboration30
Communication and collaboration
Decision log2
Decision logArchitecture decision record
Facilitate group decisions3
Facilitate group decisionsFacilitate a system design reviewLead a retrospective
Negotiation6
NegotiationBest alternative to a negotiated agreementNegotiate an employment offerNegotiate engineering trade-offsReservation valueSalary and employment-offer negotiation
Collaborate across functionsCommunicate uncertaintyConduct a one-on-oneCoordinate asynchronouslyExpectations alignmentExplain technical mechanismsGive actionable feedbackListen and clarifyManage stakeholder relationshipsOne-on-one agendaOne-on-one meetingPersuade with evidenceResolve work-related conflictReview recordStakeholder mapTechnical presentationUse feedback to improveWrite technical explanations
Leadership and management19
Leadership and managementCalibrate engineering performanceCareer development conversationCoach problem solvingDelegate engineering workEstablish system ownershipLead engineering changeLeadership without authorityManage operational workloadManage organizational dependenciesManage tasks and commitmentsMentor an engineerOnboard an engineerPerformance reviewPlan team capabilitiesResponsibility matrixSponsor professional growthTeam charterTeam profile
Keeping up with ML innovation1018
Keeping up with ML innovation
Computer vision20
Computer vision
Image reconstruction2
Image reconstructionImage denoising
Camera calibrationDepth estimationDiagnose a vision objectiveEvaluate a vision model across conditionsImage classificationImage generationImage registrationInstance segmentationIntersection over unionMultiscale feature extractionNon-maximum suppressionObject detectionObject trackingOptical character recognitionOptical flow estimationPose estimationSemantic segmentation
Conference0
Scientific conference0
Foundations for ML291
Foundations for ML
Complex systems31
Complex systems
Proxy metric failure3
Proxy metric failureResume keyword stuffingReward hacking
Sociotechnical systems4
Sociotechnical systems
Organizational incentives2
Organizational incentivesAudit incentive alignment
Audit incentive alignmentInformation asymmetryOrganizational incentives
Systems thinking10
Systems thinking
Delayed feedback2
Delayed feedbackDelayed labels
Feedback loop2
Feedback loopPerformative prediction
Analyze system-level trade-offsLimits of observabilityMap an ML feedback systemSystem boundary mappingSystem map
Proxy objective2
Proxy objectiveGoodhart-style proxy failure
System coupling2
System couplingHidden coupling
Assumption registerChallenge engineering assumptionsEmergent behaviorGoodhart-style proxy failureHidden couplingIrreversibilityMake decisions under uncertaintyPath dependencePremortemProxy objectiveReversible decision makingScenario analysisSystem coupling
Computer science79
Computer science
Algorithms17
Algorithms
Sliding window2
Sliding windowMonotonic queue
Algorithmic randomizationBacktrackingBinary searchBit manipulationBreadth-first searchDepth-first searchDynamic programmingGreedy algorithmsImplement and explain core algorithmsPrefix sumsRecursionSortingTopological sortingTwo pointers
Computational complexity3
Computational complexityAmortized analysisFree-list allocator
Computer networking10
Computer networking
Network protocol7
Network protocolHTTPInternet ProtocolSSHTCPTLSUDP
DNSRemote direct memory access
Concurrency5
ConcurrencyAsynchronous executionCondition variableConsistency modelReaders-writer lock
Data structures18
Data structures
B-tree2
B-treeB+ tree
Vector index4
Vector indexBall tree indexInverted-file vector indexk-d tree index
ArrayDirected acyclic graphHash tableHeapLinked listLog-structured merge storageMapping abstract data typeQueueStackTrieUnion-find
Distributed systems16
Distributed systems
Sharding3
ShardingDataset shardingParameter sharding
CAP theoremConsensusConsistency modelConsistent hashingDiagnose a distributed failureDistributed task executionIdempotencyIdempotency keyLeaseLineage-based recoveryRemote procedure callReplication
Operating systems6
Operating systems
Operating system process2
Operating system processData-loader worker processes
FilesystemOperating system threadVirtual memory
Invariant2
InvariantSynchronous update
Deterministic executionFinite-state machineInvariantSynchronous update
Mathematics107
Mathematics
Calculus8
CalculusChain ruleDerive and check gradientsDirectional derivativeGradientHessianJacobianPartial derivative
Control theory2
Control theoryModel predictive control
Graph theory3
Graph theoryDirected acyclic graphShortest-path computation
Information theory6
Information theoryCross-entropyEntropyInformation gainKullback–Leibler divergenceMutual information
Linear algebra19
Linear algebra
Vector spaces7
Vector spacesLinear combinationLinear independenceLinear spanLinear subspaceVector-space basisVector-space dimension
Cosine similarityEigendecompositionImplement and explain matrix operationsInner productLow-rank approximationMatrix multiplicationMatrix rankPositive semidefinite matrixSingular value decompositionTensor algebraVector norm
Numerical computing14
Numerical computing
Automatic differentiation5
Automatic differentiation
Reverse-mode automatic differentiation4
Reverse-mode automatic differentiation
Backpropagation2
BackpropagationVanishing gradient
Vector–Jacobian product
Floating-point arithmetic4
Floating-point arithmeticbfloat16 (BF16)FP16 (half precision)FP8
bfloat16 (BF16)Diagnose numerical failureFloating-point arithmeticFP16 (half precision)FP8Log-sum-exp stabilizationNumerical conditioningNumerical stability
Optimization28
Optimization
Constrained optimization6
Constrained optimizationDuality gapKarush–Kuhn–Tucker conditionsL-BFGS-BLagrange multipliersTrajectory optimization
Convex optimization2
Convex optimizationSubgradient
Gradient descent8
Gradient descent
Stochastic gradient descent6
Stochastic gradient descent
Adaptive learning-rate method5
Adaptive learning-rate method
Adam2
AdamAdamW
AdaGradRMSProp
Momentum optimization
Quasi-Newton method3
Quasi-Newton methodBFGSL-BFGS-B
Robust optimization2
Robust optimizationAdversarial training
Backtracking line searchBayesian optimizationDiagnose optimization convergenceNewton method for optimizationProximal gradient methodProximal operatorTrust-region method
Probability theory26
Probability theory
Probability distribution7
Probability distributionBernoulli distributionBeta distributionBinomial distributionGeometric distributionNormal distributionPoisson distribution
Stochastic processes5
Stochastic processes
Queuing theory2
Queuing theoryLittle's law
NonstationarityStationarity
Stochasticity2
StochasticityStochastic outcome
Base rateBayes' ruleCentral limit theoremConditional independenceConditional probabilityCovarianceExpectationExplain conditional uncertaintyLaw of large numbersRandom variableStochastic outcomeStochasticityVariance
Statistics74
Statistics
Causal inference25
Causal inference
Causal effect estimation6
Causal effect estimationDifference in differencesDoubly robust estimationInstrumental variablesInverse probability weightingRegression discontinuity
Randomized experiment6
Randomized experiment
A/B test3
A/B testCUPEDSample ratio mismatch
Ratio metricSwitchback experiment
Interference between units2
Interference between unitsExperiment carryover
Average treatment effectCausal graphCausal identificationConfoundingDistinguish prediction from causal claimsDouble machine learningExperiment carryoverInterference between unitsPotential outcomesPropensity scoreSynthetic controlUplift modeling
Sources of uncertainty7
Sources of uncertainty
Computational nondeterminism2
Computational nondeterminismNumerical nondeterminism
Epistemic uncertainty2
Epistemic uncertaintyModel uncertainty
Aleatoric uncertaintyEpistemic uncertaintyModel uncertaintyRandom-seed sensitivity
Statistical inference25
Statistical inference
Hypothesis testing6
Hypothesis testingP-valuePermutation testPower analysisSequential testingStatistical power
Selection bias2
Selection biasExposure bias in recommendation
Statistical estimation12
Statistical estimation
Maximum likelihood estimation2
Maximum likelihood estimationExpectation-maximization
Standard error2
Standard errorDelta method
Sampling distribution1
Sampling distribution
Bootstrap resamplingConfidence intervalEstimator biasEstimator consistencyEstimator efficiencyMaximum a posteriori estimationSampling distribution
Design a statistical analysisEffect sizeMultiple testingPractical significance
Survival analysis3
Survival analysisKaplan-Meier estimatorRight censoring
Uncertainty quantification4
Uncertainty quantificationConformal predictionConstruct and audit prediction setsPrediction interval
Measurement error2
Measurement errorLabel noise
Missing data mechanisms2
Missing data mechanismsAnalyze and handle missing data
Analyze and handle missing dataCoverageLabel noiseMeasurement errorMissing data mechanismsProbabilistic modelingRobust statisticsSpearman rank correlationTime series
Machine learning480
Machine learning
Automated machine learning5
Automated machine learningHyperparameter optimizationModel-search budgetModel-selection search spaceNeural architecture search
Classical machine learning16
Classical machine learning
Ensemble learning6
Ensemble learning
Bagging2
BaggingOut-of-bag evaluation
Boosting2
BoostingAdaptive boosting
Stacking
Kernel methods8
Kernel methods
Gaussian process2
Gaussian processGaussian-process covariance kernel
Support vector machine4
Support vector machineOne-class SVMSupport vector classificationSupport vector regression
Quantum feature kernel
Build a defensible tabular baseline
Continual learning3
Continual learningCatastrophic forgettingEvaluate continual learning and retention
Deep learning83
Deep learning
Neural network80
Neural network
Activation function3
Activation functionGaussian error linear unitRectified linear unit
Attention18
Attention
Multi-head attention2
Multi-head attentionGrouped-query attention
Self-attention2
Self-attentionInvariant point attention
Soft attention8
Soft attention
Dot-product attention3
Dot-product attention
Scaled dot-product attention2
Scaled dot-product attentionFlashAttention
Additive attentionGraves Gaussian-window attentionLuong global attentionLuong local attention
Cross-attentionGraph attentionHard attentionImplement and explain attentionSliding-window attention
Autoencoder4
AutoencoderMasked autoencoderSparse autoencoderVariational autoencoder
Convolutional neural network14
Convolutional neural network
Faster R-CNN4
Faster R-CNN
Mask R-CNN2
Mask R-CNNRoIAlign
Region proposal network
AlexNetEfficientNetInception networkLeNetMobileNetRegion proposal networkResidual networkU-NetVGG networkYOLO detector family
Encoder-decoder architecture3
Encoder-decoder architectureSequence-to-sequence modelingU-Net
Graph neural network5
Graph neural networkGraph attentionGraph convolutionGraph message passingGraphSAGE
Neural recommendation model7
Neural recommendation modelBERT4RecDeep learning recommendation modelDeepFMNeural collaborative filteringSASRecWide & Deep model
Normalization layer4
Normalization layerBatch normalizationLayer normalizationRMS normalization
Recurrent neural network4
Recurrent neural networkGraphRNNLong short-term memory networkRecurrent visual attention model
Transformer12
Transformer
Positional encoding5
Positional encoding
Rotary position embedding3
Rotary position embeddingImplement and inspect RoPERoPE scaling (position interpolation and YaRN)
Attention with linear biases
Vision transformer2
Vision transformerSwin Transformer
BERTBERT4RecDETRSASRec
AlphaFoldCLIPDeep Q-networkMambaMixture of expertsMultilayer perceptronResidual connectionTrain and diagnose a neural networkTwo-tower retrieval model
Geometric deep learningNeural scaling laws
Generalization14
Generalization
Distribution shift7
Distribution shiftAssess and respond to distribution shiftConcept driftCovariate shiftFeature driftLabel shiftTraining-serving skew
Bias–variance trade-offDouble descentOverfittingProbably approximately correct learningStatistical-query learningUnderfitting
Graph learning14
Graph learning
Graph generation2
Graph generationMolecular graph generation
Link prediction2
Link predictionKnowledge graph completion
Evaluate graph generationGraph classificationGraph neural networkGraph oversmoothingInductive graph learningKnowledge graphNode classificationTrain and evaluate a graph modelTransductive graph learning
Learning theory5
Learning theory
Probably approximately correct learning2
Probably approximately correct learningStatistical-query learning
Probably approximately correct learningQuery learningStatistical-query learningVapnik–Chervonenkis dimension
Loss function63
Loss function
Classification loss family10
Classification loss family
Cross-entropy loss4
Cross-entropy lossBinary cross-entropy lossLabel-smoothed cross-entropySoftmax ranking loss
Brier lossFocal lossHinge lossSquared hinge lossZero-one loss
Intersection-over-union loss2
Intersection-over-union lossGeneralized IoU loss
Negative log-likelihood loss6
Negative log-likelihood lossConnectionist temporal classification lossCRF negative log-likelihood lossGaussian negative log-likelihood lossListMLE lossPoisson negative log-likelihood loss
Ranking loss family9
Ranking loss familyApproximate NDCG lossBayesian personalized ranking objectiveListMLE lossListNet lossPairwise hinge ranking lossPairwise logistic ranking lossSelect and audit a ranking objectiveSoftmax ranking loss
Reconstruction loss family4
Reconstruction loss familyDenoising reconstruction lossMasked reconstruction lossPerceptual feature loss
Regression loss family12
Regression loss familyAbsolute-error lossCRPS training lossEpsilon-insensitive lossExpectile lossGaussian negative log-likelihood lossHuber lossLog-cosh lossPoisson negative log-likelihood lossQuantile lossSquared-error lossTweedie deviance loss
Representation learning loss family9
Representation learning loss familyBarlow Twins objectiveContrastive margin lossCosine embedding lossInfoNCE lossNormalized temperature-scaled cross-entropy lossSupervised contrastive lossTriplet lossVICReg objective
Structured prediction loss family4
Structured prediction loss familyConnectionist temporal classification lossCRF negative log-likelihood lossStructured hinge loss
Surrogate loss4
Surrogate lossApproximate NDCG lossPPO clipped surrogate objectiveStructured hinge loss
Cox partial-likelihood lossDenoising diffusion lossDiagnose a vision objectiveDice overlap lossGAN logistic lossKL-divergence lossNegative ELBO lossSelect and validate a loss functionTemporal-difference fitting lossTversky loss
Machine learning for software engineering3
Machine learning for software engineeringCode generationSemantic code search
Model training45
Model training
Decision-tree induction15
Decision-tree induction
Tree split criterion8
Tree split criterion
Classification node impurity2
Classification node impurityGini impurity
Compute and explain tree split criteriaGain ratioInformation gainPoisson-deviance split criterionSquared-error split criterion
C4.5 tree inductionCART tree inductionID3 tree inductionRecursive binary splittingSparsity-aware split findingTree split-selection bias
Gradient-based training10
Gradient-based training
Learning-rate schedule3
Learning-rate scheduleCosine learning-rate decayLearning-rate warmup
Activation checkpointingCompare optimization methodsGradient accumulationGradient clippingMixed precision trainingWeight initialization
Knowledge distillation2
Knowledge distillationDistilled reasoning training
Regularization13
Regularization
Data augmentation2
Data augmentationDomain randomization
Decision-tree pruning2
Decision-tree pruningCost-complexity pruning
DropoutEarly stoppingGradient-norm penaltyL1 regularizationL2 regularizationPolicy entropy regularizationTree growth stoppingWeight decay
Adversarial minimax trainingEmpirical risk minimizationSelf-supervised pretrainingTraining epoch
Multimodal learning6
Multimodal learning
Document AI4
Document AI
Document understanding2
Document understandingForm-field extraction
Optical character recognition
CLIP
Probabilistic modeling43
Probabilistic modeling
Bayesian inference9
Bayesian inference
Bayesian model checking5
Bayesian model checkingPosterior predictive checkingPosterior sensitivity analysisPrior predictive checkingSimulation-based calibration
Fit and diagnose a Bayesian modelKalman filterMonte Carlo dropout
Markov chain Monte Carlo5
Markov chain Monte Carlo
Hamiltonian Monte Carlo2
Hamiltonian Monte CarloNo-U-Turn sampler
Effective sample sizeR-hat convergence diagnostic
Probabilistic model20
Probabilistic model
Generative model8
Generative model
Diffusion model2
Diffusion modelLatent diffusion model
Deep Boltzmann machineGenerative adversarial networkNormalizing flowScore-based generative modelVariational autoencoder
Hierarchical Bayesian model3
Hierarchical Bayesian model
Latent Dirichlet allocation2
Latent Dirichlet allocationSupervised latent Dirichlet allocation
Probabilistic graphical model5
Probabilistic graphical model
State-space model3
State-space modelHidden Markov modelMamba
Deep Boltzmann machine
Beta–binomial Bayesian modelGaussian mixture modelGaussian processNaive Bayes
Variational inference2
Variational inferenceReparameterization trick
Posterior distribution2
Posterior distributionPosterior predictive distribution
Prior distribution2
Prior distributionConjugate prior
Conjugate priorPosterior distributionPosterior predictive distributionPrior distributionProbabilistic programmingStatistical model identifiability
Quantum machine learning6
Quantum machine learningQuantum feature kernelQuantum feature mapQuantum measurement for learningQuantum state preparationVariational quantum classifier
Reinforcement learning42
Reinforcement learning
Multi-armed bandit2
Multi-armed banditContextual bandit
Policy gradient8
Policy gradient
Actor–critic3
Actor–criticDeep deterministic policy gradientSoft actor-critic
Proximal policy optimization2
Proximal policy optimizationGroup relative policy optimization
Deterministic policy gradientREINFORCE
Reinforcement-learning training environment2
Reinforcement-learning training environmentReinforcement-learning reward specification
Temporal-difference learning3
Temporal-difference learning
Q-learning2
Q-learningDeep Q-network
Model-based reinforcement learning2
Model-based reinforcement learningMonte Carlo tree search
Offline reinforcement learning2
Offline reinforcement learningAudit an offline policy estimate
Audit an offline policy estimateBellman equationEpisode terminationEpisode truncationExperience replayExploration–exploitation trade-offImitation learningMarkov decision processMaximum-entropy reinforcement learningModel-based reinforcement learningMonte Carlo return estimationMonte Carlo tree searchOff-policy evaluationOffline reinforcement learningPartially observable Markov decision processPolicyPolicy optimizationReinforcement learning from human feedbackReinforcement learning with verifiable rewardsReward hackingTarget networkThompson samplingTrain and evaluate a DQN agentUpper confidence boundValidate an RL training environmentValue function
Representation learning12
Representation learning
Contrastive learning3
Contrastive learningHard-negative miningNegative sampling
Embedding7
Embedding
Word embedding3
Word embeddingContinuous bag-of-words modelSkip-gram model
Contextual embeddingEmbedding modelGraph embedding
Metric learning
Supervised learning72
Supervised learning
Classification12
Classification
Image classification2
Image classificationCrop disease recognition from images
Binary classificationClick-through rate predictionGraph classificationMalware classificationMulticlass classificationMultilabel classificationNode classificationOrdinal predictionText classification
Learning to rank7
Learning to rank
Pairwise ranking4
Pairwise ranking
LambdaRank2
LambdaRankLambdaMART
Bradley-Terry model
Listwise rankingPointwise ranking
Model prediction4
Model predictionPredicted class labelPredicted class probabilityPredicted numerical value
Prediction target variable5
Prediction target variable
Numerical target3
Numerical targetContinuous targetCount target
Categorical targetContinuous targetCount targetNumerical target
Predictive model37
Predictive model
Decision tree12
Decision tree
Decision-tree internal node2
Decision-tree internal nodeDecision-tree split
Axis-aligned decision treeClassification treeConditional inference treeDecision stumpDecision-tree leafExplain a tree predictionFit, prune and validate decision treesOblique decision treeRegression tree
Generalized linear model6
Generalized linear model
Linear regression3
Linear regressionElastic netRidge regression
Logistic regressionPoisson regression
Tree ensemble8
Tree ensemble
Random forest3
Random forestQuantile regression forestRandom feature subsampling at splits
Compare trees, bagging, forests and boostingExtremely randomized treesGradient-boosted treesIsolation forest
System One models, TypeSafe branding1
Jev
JevK-nearest neighborsNaive BayesOutcome reward modelProcess reward modelReward modelSelect a model using evidenceSupport vector machineSurrogate modelVariational quantum classifier
Regression2
RegressionQuantile regression
Active learningClass imbalanceCost-sensitive learningPreference learning
Time series17
Time series
ARIMA2
ARIMASeasonal ARIMA
Time-series forecasting7
Time-series forecastingBuild a forecasting baselineDemand forecastingEnergy load forecastingRenewable power forecastingTraffic forecastingWeather forecasting
AutocorrelationChange-point detectionDifferencingExponential smoothingForecast reconciliationSeasonalityState-space model
Training objective9
Training objective
Adversarial generative training objective3
Adversarial generative training objectiveGAN logistic lossWasserstein critic objective
Composite training objective2
Composite training objectiveVICReg objective
Adversarial generative training objectiveComposite training objectiveEvidence lower boundGAN logistic lossLoss reduction and weightingPPO clipped surrogate objectiveVICReg objectiveWasserstein critic objective
Transfer learning10
Transfer learning
Fine-tuning7
Fine-tuning
Parameter-efficient fine-tuning3
Parameter-efficient fine-tuning
Low-rank adaptation2
Low-rank adaptationQLoRA
Adapt a pretrained modelInstruction tuningSupervised reasoning fine-tuning
Multi-task learningSimulation-to-real transfer
Unsupervised learning21
Unsupervised learning
Anomaly detection5
Anomaly detectionEvaluate an anomaly detectorNovelty detectionOut-of-distribution detectionOutlier detection
Clustering5
ClusteringDBSCANHierarchical clusteringK-meansSpectral clustering
Density estimation2
Density estimationScore matching
Dimensionality reduction6
Dimensionality reduction
Principal component analysis2
Principal component analysisKernel principal-component analysis
Interpret dimensionality reduction responsiblyt-SNEUMAP
Self-supervised learning2
Self-supervised learningSelf-supervised pretraining
Hyperparameter2
HyperparameterTraining batch size
Meta-learning2
Meta-learningModel-agnostic meta-learning
Model parameter2
Model parameterParameter sparsity
HyperparameterMeta-learningModel inferenceModel parameterModel-agnostic meta-learningOnline learningParameter sparsitySemi-supervised learningTraining batch size
Natural language processing74
Natural language processing
Autoregressive language model5
Autoregressive language model
Large language model4
Large language modelContext windowHallucinationReasoning model
Bag of words2
Bag of wordsTF–IDF
Decoding strategy5
Decoding strategyBeam searchNucleus samplingTemperature samplingTop-k sampling
Information extraction4
Information extractionForm-field extractionNamed entity recognitionResume parsing
Language modeling2
Language modelingMasked language modeling
Machine translation3
Machine translationBLEUNeural machine translation
Preference optimization9
Preference optimizationConstitutional AIDirect preference optimizationKahneman-Tversky optimization (KTO)Odds ratio preference optimization (ORPO)Reinforcement learning from human feedbackRejection sampling fine-tuningReward modelSimple preference optimization (SimPO)
Prompt engineering8
Prompt engineering
Prompt optimization3
Prompt optimization
GEPA2
GEPAInstance-wise Pareto prompt selection
Chain-of-thought promptingIn-context learningSelf-consistencyTree of thoughts
Reasoning training6
Reasoning trainingDistilled reasoning trainingOutcome supervisionProcess supervisionReinforcement learning with verifiable rewardsSupervised reasoning fine-tuning
Retrieval-augmented generation5
Retrieval-augmented generationAgentic RAGDocument chunkingGraphRAGRAG evaluation
Structured generation2
Structured generationConstrained decoding
Test-time compute scaling4
Test-time compute scalingBest-of-N samplingSelf-consistencyTree of thoughts
Tokenization4
TokenizationByte-pair encodingUnigram tokenizationWordPiece tokenization
Topic model3
Topic modelCorrelated topic modelLatent Dirichlet allocation
Automatic speech recognitionBuild a text-classification baselineCompare decoding policiesDocument AIPerplexityReasoning traceROUGESequence generationSummarizationText classificationText-to-speech synthesisTopic modelingTrain and audit a tokenizer
Research practice157
Research practice
Literature review and synthesis4
Literature review and synthesisConduct a literature reviewDevelop an evidence-based book argumentRead a research paper critically
ML research community1
ML research community
Classical machine learning researchers0
Computer vision researchers0
Deep learning foundations researchers0
Generative modeling researchers0
Language and sequence modeling researchers0
Learning theory researchers0
ML systems researchers0
Probabilistic modeling researchers0
Reinforcement learning researchers0
Statistics and causal inference researchers0
Trustworthy ML researchers0
Reproducibility and replication5
Reproducibility and replicationAudit experiment variabilityMaintain research provenanceReproduce a research resultResearch reproduction
Research investigation2
Research investigationReplication study
Research paper126
Research paper
Classical machine learning paper20
Classical machine learning paper
Decision tree and ensemble paper11
Decision tree and ensemble paperA Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting (1997)A System for Induction of Oblique Decision Trees (1994)Bagging Predictors (1996)Extremely randomized trees (2006)Greedy Function Approximation: A Gradient Boosting Machine (2001)Induction of Decision Trees (1986)Quantile Regression Forests (2006)Random Forests (2001)Unbiased Recursive Partitioning: A Conditional Inference Framework (2006)XGBoost: A Scalable Tree Boosting System
Distance Metric Learning for Large Margin Nearest Neighbor ClassificationEstimating the Support of a High-Dimensional Distribution (2001)Learning a Distance Metric from Relative ComparisonsNonlinear Component Analysis as a Kernel Eigenvalue Problem (1998)On Spectral Clustering: Analysis and an Algorithm (2001)Regression Shrinkage and Selection via the Lasso (1996)Regularization Paths for Generalized Linear Models via Coordinate Descent (2010)Support-vector networks (1995)
Computer vision paper12
Computer vision paperDeep Residual Learning for Image RecognitionFaceNet: A Unified Embedding for Face Recognition and ClusteringFaster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksFocal Loss for Dense Object DetectionGoing Deeper with ConvolutionsGradient Based Learning Applied to Document RecognitionImageNet Classification with Deep Convolutional Neural NetworksImageNet Large Scale Visual Recognition ChallengeMask R-CNN (publication)Recurrent Models of Visual AttentionVery Deep Convolutional Networks for Large-Scale Image Recognition
Generative modeling paper7
Generative modeling paperAuto-Encoding Variational BayesDeep Boltzmann MachinesDenoising Diffusion Probabilistic ModelsGenerative Adversarial NetworksGenerative Modeling by Estimating Gradients of the Data DistributionThe Helmholtz Machine (1995)
Language and sequence modeling paper15
Language and sequence modeling paperA Decomposable Attention Model for Natural Language InferenceA Structured Self-attentive Sentence EmbeddingAttention Is All You NeedBERT: Pre-training of Deep Bidirectional Transformers for Language UnderstandingDistributed Representations of Words and Phrases and their CompositionalityEffective Approaches to Attention-based Neural Machine TranslationEfficient Estimation of Word Representations in Vector SpaceGenerating Sequences With Recurrent Neural NetworksLong Short-Term MemoryLong Short-Term Memory-Networks for Machine ReadingNeural Machine Translation by Jointly Learning to Align and TranslateNeural Machine Translation of Rare Words with Subword UnitsRoFormer: Enhanced Transformer with Rotary Position EmbeddingSequence to Sequence Learning with Neural Networks
Large language model paper5
Large language model paperDeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language ModelsDSPy: Compiling Declarative Language Model Calls into Self-Improving PipelinesGEPA: Reflective Prompt Evolution Can Outperform Reinforcement LearningJudging LLM-as-a-Judge with MT-Bench and Chatbot Arena
Learning theory paper6
Learning theory paperA Theory of the Learnable (1984)Efficient Noise-Tolerant Learning from Statistical Queries (1998)Learning from Noisy Examples (1988)On the Uniform Convergence of Relative Frequencies of Events to Their Probabilities (1971)Queries and Concept Learning (1988)
ML systems paper4
ML systems paperRay: A Distributed Framework for Emerging AI ApplicationsResilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster ComputingTensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Optimization and training paper9
Optimization and training paperA Stochastic Approximation MethodAdam: A Method for Stochastic OptimizationBatch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate ShiftDecoupled Weight Decay RegularizationDropout: A Simple Way to Prevent Neural Networks from OverfittingImproving neural networks by preventing co-adaptation of feature detectorsLearning representations by back-propagating errorsTaylor expansion of the accumulated rounding error
Probabilistic modeling and inference paper14
Probabilistic modeling and inference paperAn Introduction to Variational Methods for Graphical Models (1999)Bayesian Interpolation (1992)Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep LearningDynamic Bayesian Networks: Representation, Inference and Learning (2002)Gaussian Processes for Regression (1995)Inference from Iterative Simulation Using Multiple Sequences (1992)Latent Dirichlet Allocation (2003)Loopy Belief Propagation for Approximate Inference: An Empirical Study (1999)MCMC Using Hamiltonian Dynamics (2011)Posterior Predictive Assessment of Model Fitness via Realized Discrepancies (1996)Stochastic Variational Inference (2013)Variational Inference with Normalizing FlowsVariational Principal Components (1999)
Reinforcement learning paper13
Reinforcement learning paperAdaptive Exploration in Linear Contextual Bandit (2019)Computing Factored Value Functions for Policies in Structured MDPs (1999)Continuous control with deep reinforcement learningLearning to Predict by the Methods of Temporal Differences (1988)Learning with Good Feature Representations in Bandits and in RL with a Generative Model (2020)Neuronlike Adaptive Elements That Can Solve Difficult Learning Control Problems (1983)On the Theory of Dynamic Programming (1952)Playing Atari with Deep Reinforcement LearningPolicy Gradient Methods for Reinforcement Learning with Function Approximation (1999)Proximal Policy Optimization AlgorithmsQ-learning (1992)Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
Statistics and causal inference paper8
Statistics and causal inference paperAssessing Sensitivity to an Unobserved Binary Covariate in an Observational Study with Binary Outcome (1983)Better Bootstrap Confidence Intervals (1987)Bootstrap Methods: Another Look at the Jackknife (1979)Causal Diagrams for Empirical Research (1995)Estimation of Regression Coefficients When Some Regressors Are Not Always Observed (1994)Inference and Missing Data (1976)The Central Role of the Propensity Score in Observational Studies for Causal Effects (1983)
Trustworthy ML paper7
Trustworthy ML paperCalibrating Noise to Sensitivity in Private Data AnalysisEquality of Opportunity in Supervised LearningModel Cards for Model ReportingSemi-supervised Knowledge Transfer for Deep Learning from Private Training DataTowards Deep Learning Models Resistant to Adversarial AttacksTowards Evaluating the Robustness of Neural Networks
A New Algorithm for Data CompressionEfficient and robust approximate nearest neighbor search using Hierarchical Navigable Small World graphsHighly accurate protein structure prediction with AlphaFoldModel-Agnostic Meta-Learning for Fast Adaptation of Deep NetworksRepresentation Learning with Contrastive Predictive Coding
Research record8
Research recordClaim-evidence ledgerCompute budget reportExperiment configurationLiterature mapPredeclared research protocolReproduction packageResearch log
Scholarly book6
Scholarly bookAn Introduction to Computational Learning Theory (1994)C4.5: Programs for Machine Learning (1993)Classification and Regression Trees (1984)Gaussian Processes for Machine Learning (2006)The Algorithmic Foundations of Differential Privacy
Design an informative experimentIntellectual honestyResearch peer reviewReview research constructively
Robotics12
RoboticsDigital twinMotion planningRobot controlRobot kinematicsRobot localizationRobot perceptionRobot safety envelopeSensor fusionSimulation-to-real transferSimultaneous localization and mappingValidate a robot policy in simulation
Designing production ML systems999
Designing production ML systems
Computing infrastructure56
Computing infrastructure
Cloud computing20
Cloud computing
Managed service16
Managed serviceAmazon Elastic Kubernetes ServiceAmazon EMRAmazon Managed Workflows for Apache AirflowAmazon SageMaker AIAzure DatabricksAzure Kubernetes ServiceAzure Machine LearningCloud ComposerDataflowDataprocGoogle Kubernetes EngineSageMaker CanvasSageMaker managed MLflowVertex AI PipelinesVertex AI platform family
Amazon Web Services platformGoogle CloudMicrosoft Azure
Cluster workload manager7
Cluster workload managerAmazon Elastic Kubernetes ServiceAzure Kubernetes ServiceGoogle Kubernetes EngineHadoop YARNKubernetesSlurm
Container runtime2
Container runtimeDocker
Job scheduler11
Job schedulerAmazon Managed Workflows for Apache AirflowApache AirflowApache OozieCloud ComposerCron schedulingDagsterFlyteKubeflow PipelinesPrefectVertex AI Pipelines
Message broker6
Message brokerApache ActiveMQApache KafkaApache PulsarApache QpidApache RocketMQ
Storage system6
Storage system
Filesystem2
FilesystemHDFS
Object storage2
Object storageApache Ozone
Apache BookKeeper
Web server7
Web serverApache APISIXApache GuacamoleApache HTTP ServerApache KnoxApache TomcatApache Traffic Server
Apache ZooKeeperContainer image
Data engineering136
Data engineering
Data format13
Data format
Columnar data format4
Columnar data formatApache ArrowApache ORCApache Parquet
Apache AvroApache HudiApache IcebergApache ThriftCSVGGUFJSONProtocol Buffers
Data integration13
Data integration
Data integration tool9
Data integration toolApache CamelApache FlumeApache HopApache NiFiApache SeaTunnelApache Sqoopdbt CoreSQLMesh
Change data captureExtract load transformExtract transform load
Data labeling6
Data labelingAudit delayed outcome labelsInter-annotator agreementLabel maturation windowLabeling guidelineWeak supervision
Data manipulation library5
Data manipulation libraryHugging Face DatasetsNumPypandasPolars
Data modeling3
Data modelingAnalytics semantic layerSchema evolution
Data pipeline development2
Data pipeline developmentImplement a testable data pipeline
Data processing framework12
Data processing framework
Apache Hadoop3
Apache HadoopHadoop MapReduceHadoop YARN
Apache Spark2
Apache SparkResilient distributed dataset
Apache BeamApache FlinkApache PigApache SedonaApache StormApache TezHadoop MapReduce
Data quality5
Data qualityData validationDeduplicationFeature freshnessImprove dataset quality with evidence
Database systems33
Database systems
Database20
Database
Analytical database7
Analytical databaseApache DorisApache DruidApache KuduApache KylinApache PinotDuckDB
In-memory data store6
In-memory data store
Redis2
RedisRedis sorted set
Apache GeodeApache IgniteMemcached
Apache AccumuloApache CassandraApache CouchDBApache HBaseApache IoTDBPostgreSQL
Query engine7
Query engineApache CalciteApache DataFusionApache DrillApache HiveApache ImpalaApache Phoenix
SQL2
SQLAnalytical SQL
Analytical SQLSQLSQL query performance and operating evidenceVector databaseWrite-ahead log
Dataset5
DatasetEvaluation scenario suiteImageNetIterable datasetMap-style dataset
Document processing library4
Document processing libraryApache PDFBoxApache POIApache Tika
Feature engineering11
Feature engineering
Categorical encoding4
Categorical encoding
Target encoding2
Target encodingOut-of-fold target encoding
One-hot encoding
Engineer prediction-time-valid featuresFeature hashingFeature scalingFeature selectionImputationPoint-in-time join
Stream processing14
Stream processing
Windowed aggregation4
Windowed aggregationSession windowSliding event-time windowTumbling window
Allowed lateness policyDesign an event-time pipelineEvent-time processingEvent-time triggerExactly-once effect semanticsLate-arriving dataProcessing timeStream-state checkpointWatermark
Workflow orchestration3
Workflow orchestrationDesign a recoverable workflowML workflow orchestration
Data serialization1
Data serialization
Data-centric AI2
Data-centric AISynthetic data
Batch processingData serializationData-centric AIDataset curationSparse feature representationsSynthetic data
Evaluation and experimentation116
Evaluation and experimentation
Decision-threshold selection2
Decision-threshold selectionSelective prediction
Evaluation metric48
Evaluation metric
Classification metric12
Classification metric
Precision2
PrecisionPrecision at k
Recall2
RecallRecall at k
Balanced accuracyClassification error rateExpected calibration errorF1 scoreMatthews correlation coefficientPrecision-recall AUCROC AUC
Efficiency metric14
Efficiency metric
Latency5
LatencyInter-token latencyLatency budgetTail latencyTime to first token
Throughput3
ThroughputModel FLOPs utilization (MFU)Tokens per second
Cost per successful taskEnergy per inferenceGoodputIndex build timeIndex memory footprint
Proper scoring rule6
Proper scoring rule
Strictly proper scoring rule4
Strictly proper scoring ruleBrier scoreContinuous ranked probability scoreLogarithmic score
Interval score
Ranking metric8
Ranking metric
Recall at k2
Recall at kExact-neighbor recall at k
Mean average precisionMean reciprocal rankNormalized discounted cumulative gainPrecision at kRecommendation diversity
Regression metric5
Regression metricCoefficient of determinationMean absolute errorMean squared errorRoot mean squared error
Filter selectivityGroundedness and faithfulnessGuardrail metricspass@k
Evaluation validity threat10
Evaluation validity threat
Data leakage3
Data leakageBenchmark contaminationTime-series leakage
LLM judge bias4
LLM judge biasJudge position biasJudge self-enhancement biasJudge verbosity bias
Incomplete evaluation runOffline-online evaluation gap
Experiment design2
Experiment designAblation study
LLM and agent evaluation18
LLM and agent evaluation
LLM application evaluation6
LLM application evaluationHumanEval benchmarkMMLU benchmarkNeedle-in-a-haystack testRULER benchmarkSWE-bench
LLM evaluation harness6
LLM evaluation harnessArize PhoenixDeepEvalEleutherAI LM Evaluation HarnesspromptfooRagas
Agent trajectoryAudit agent trajectoriesAudit an LLM evaluationRAG evaluationTask-success verifier
Model evaluation23
Model evaluation
Error analysis2
Error analysisAgent trajectory analysis
Held-out validation7
Held-out validation
Cross-validation2
Cross-validationNested cross-validation
Temporal validation2
Temporal validationRolling-origin evaluation
Group-aware validationTraining validation and test split
LLM-as-judge evaluation4
LLM-as-judge evaluationPairwise LLM judgingReference-guided LLM judgingSingle-answer LLM grading
Pairwise preference evaluation3
Pairwise preference evaluationElo and arena ratingsPairwise LLM judging
Ranking evaluation2
Ranking evaluationInterleaving evaluation
Calibration evaluationHuman evaluationScenario-based comparisonSlice evaluationStress testing
Sensitivity analysis2
Sensitivity analysisPosterior sensitivity analysis
Arize AIAudit a ranking metricBuild a trustworthy scenario evaluationDesign a valid model-selection protocolEvaluate human judgmentsEvaluate probabilistic forecastsEvaluation reproducibilityEvaluation workload schedulingMetric definitionProbability calibration
Hardware and compilers54
Hardware and compilers
Compiler construction15
Compiler construction
Compiler8
CompilerApache TVMLLVMMLIROpenVINOTensorRTTriton kernel languageXLA
Intermediate representation2
Intermediate representationONNX
Just-in-time compilation2
Just-in-time compilationtorch.compile
Ahead-of-time compilationKernel fusion
Hardware interconnect5
Hardware interconnectInfiniBandNVIDIA NVLinkPCI ExpressRDMA over Converged Ethernet
Memory hierarchy4
Memory hierarchyCPU cacheHigh-bandwidth memoryPinned host memory
Parallel programming11
Parallel programmingAll-reduceAmdahl's lawCollective communicationDataflow executionGEMM (general matrix multiply)GPU kernelMemory coalescingOptimize a GPU kernelTilingVectorization
Performance engineering3
Performance engineeringAnalyze sparse-computation trade-offsProfiling
Processor11
Processor
Application-specific integrated circuit2
Application-specific integrated circuitTensor Processing Unit
Central processing unit2
Central processing unitTranslation lookaside buffer
Graphics processing unit4
Graphics processing unitGPU occupancyMulti-instance GPU (MIG)Tensor Cores
Field-programmable gate arrayNVIDIA Jetson
Roofline performance model3
Roofline performance modelArithmetic intensityMemory-bound versus compute-bound kernels
Arithmetic intensityMemory-bound versus compute-bound kernelsRoofline performance modelSelect a compute platform
ML application domains47
ML application domains
Biomedical machine learning6
Biomedical machine learning
Machine learning for drug discovery2
Machine learning for drug discoveryMolecular property prediction
Machine learning in genomics2
Machine learning in genomicsGenetic variant effect prediction
Protein structure prediction
Geospatial machine learning4
Geospatial machine learning
Machine learning for remote sensing2
Machine learning for remote sensingLand-cover classification
Weather forecasting
Machine learning for cybersecurity3
Machine learning for cybersecurityIntrusion detectionMalware classification
Machine learning for trust and safety2
Machine learning for trust and safetyContent moderation
Machine learning in agriculture2
Machine learning in agricultureCrop disease recognition from images
Machine learning in education2
Machine learning in educationKnowledge tracing
Machine learning in energy systems3
Machine learning in energy systemsEnergy load forecastingRenewable power forecasting
Machine learning in finance3
Machine learning in financeCredit risk predictionFraud detection
Machine learning in healthcare3
Machine learning in healthcareClinical risk predictionMedical image analysis
Machine learning in industrial systems5
Machine learning in industrial systemsFault diagnosisPredictive maintenanceRemaining useful life predictionVisual quality inspection
Machine learning in retail and marketplaces5
Machine learning in retail and marketplacesCustomer churn predictionCustomer lifetime value predictionDemand forecastingProduct recommendation
Machine learning in transportation3
Machine learning in transportationETA predictionTraffic forecasting
Scientific machine learning4
Scientific machine learningApproximation of PDE solutionsClimate modeling with machine learningMaterials discovery
Machine learning in advertising
ML systems121
ML systems
Agent systems14
Agent systems
Bounded tool execution2
Bounded tool executionQuery validation
Model tool use2
Model tool useModel Context Protocol
Agent harnessAgent memoryAgentic RAGBuild a bounded agent harnessContext managementLLM guardrailsMulti-agent systemsReAct (reason and act)Validate a model-proposed analytics query
Data loading7
Data loadingBatch collationData prefetchingData-loader worker processesData-loader worker seedingMini-batchingTraining-data shuffling
Distributed training17
Distributed training
Data parallelism4
Data parallelismDistributedDataParallel (DDP)Fully sharded data parallel (FSDP)ZeRO (Zero Redundancy Optimizer)
Federated learning3
Federated learningAudit federated training assumptionsFederated averaging
Collective communicationExpert parallelismMegatron-LMModel parallelismNCCL collective operationsParameter shardingPipeline parallelismSequence parallelismTensor parallelism
ML platform15
ML platform
Feature store2
Feature storeFeast
Kubeflow2
KubeflowKubeflow Katib
Amazon SageMaker AIArize platformAzure Machine LearningDVCMetaflowMichelangeloMLflowSageMaker managed MLflowTensorFlow ExtendedVertex AI platform family
ML platform engineering2
ML platform engineeringML platform development
MLOps13
MLOps
Evaluation run record2
Evaluation run recordTree-model comparison report
Model registry2
Model registryHugging Face Hub
Continuous trainingDrift detectionML workflow orchestrationModel artifactModel monitoringModel versioningOperate a model lifecycleTraining run record
Model serving50
Model serving
Dynamic batching2
Dynamic batchingContinuous batching
Edge machine learning5
Edge machine learningCore MLExecuTorchllama.cppMLX
LLM inference systems11
LLM inference systems
Prefill and decode phases3
Prefill and decode phasesChunked prefillDisaggregated prefill and decode
Speculative decoding4
Speculative decodingEAGLE decodingLookahead decodingMedusa decoding
Benchmark LLM servingKey-value cachingLLM request scheduling
Model compression3
Model compression2:4 structured sparsityEarly exit
Model quantization11
Model quantizationActivation-aware weight quantizationGPTQINT4 quantizationINT8 quantizationKV cache quantizationNF4 (4-bit NormalFloat)Post-training quantizationQuantization-aware trainingSmoothQuantWeight-only quantization
Model serving system15
Model serving systemBentoMLCore MLExecuTorchKServellama.cppONNX RuntimeOpenVINORay ServeSGLangTensorRTTensorRT-LLMText Generation InferenceTriton Inference ServervLLM
Batch versus online inferenceModel pruningModel routingServerless inferenceStreaming inference
Hugging FaceML technical debt
Product and strategy22
Product and strategy
Frame an ML problem2
Frame an ML problemDefine and justify a prediction target
ML product economics3
ML product economicsBusiness caseEvaluate build versus buy
Product discovery4
Product discoveryElicit requirementsProblem briefProduct discovery interview
Product metrics4
Product metricsEvaluate product impactMetric specificationSelect success metrics
Assess ML feasibilityDefine release criteriaDevelop engineering strategyEngineering roadmapManage the ML product lifecyclePrioritize engineering workProduct requirements documentStrategic planning
Recommender systems17
Recommender systems
Collaborative filtering3
Collaborative filteringMatrix factorization for recommendationNeural collaborative filtering
Recommendation5
RecommendationCandidate generationFeed rankingProduct recommendationSequential recommendation
Implicit feedback2
Implicit feedbackPosition bias
Cold-start recommendationContent-based recommendationDesign a recommendation pipelineExposure bias in recommendationImplicit feedbackMulti-stage rankingNeural recommendation modelPosition bias
Reliability and operations49
Reliability and operations
GPU fleet lifecycle management3
GPU fleet lifecycle managementDiagnose GPU healthGPU active diagnostics
Incident management9
Incident management
Incident postmortem review2
Incident postmortem reviewWrite an incident postmortem
On-call duty2
On-call dutyRespond while on call
Communicate during an incidentCoordinate an incident responseIncident reportRunbook
Observability10
Observability
Design operational monitoring2
Design operational monitoringOperational logging
Monitoring system5
Monitoring systemGrafanaNVIDIA Data Center GPU ManagerOpenTelemetryPrometheus
Drift detectionGPU health monitoring
Resilience pattern11
Resilience patternAdaptive concurrency limitBackpressureBulkhead isolationCircuit breakerFallback designGraceful degradationLoad sheddingRate limitingRetries with backoff and jitterRetry budget
Site reliability engineering14
Site reliability engineeringCascading failureDefine service-level objectivesError budgetOperational readiness review recordOperational toilPlan operational capacityReduce operational toilRetry amplificationService-level indicatorService-level objectiveSynchronized retry burstTest overload and recoveryValidate disaster recovery
Checkpointing
Search and retrieval systems106
Search and retrieval systems
Evidence grounding2
Evidence groundingBuild grounded generation
Information retrieval21
Information retrieval
Dense retrieval2
Dense retrievalColBERT
Reranking3
RerankingCross-encoder rerankerExact vector-distance reranking
Semantic search2
Semantic searchSemantic code search
Sparse retrieval2
Sparse retrievalBM25
Vector similarity search3
Vector similarity searchFiltered vector searchMaximum inner-product search
Candidate generationEvaluate retrieval and rerankingHybrid searchHypothetical document embeddings (HyDE)Query expansionQuery intentRecruiter searchSearch ranking
Search engine software4
Search engine softwareApache LuceneApache NutchApache Solr
Vector retrieval78
Vector retrieval
Approximate nearest-neighbor search54
Approximate nearest-neighbor search
Disk-oriented approximate neighbor search4
Disk-oriented approximate neighbor searchDiskANN search architectureFreshDiskANNSPANN disk search
Graph-based approximate neighbor search14
Graph-based approximate neighbor search
Hierarchical navigable small world search3
Hierarchical navigable small world searchHNSW-PQ searchHNSW-SQ search
ACORN filtered searchCAGRA graph searchDiskANN search architectureFiltered VamanaFreshDiskANNNavigable small-world searchNavigating spreading-out graph searchNN-Descent graph constructionStitched VamanaVamana graph search
Hash-based approximate neighbor search10
Hash-based approximate neighbor search
Locality-sensitive hashing6
Locality-sensitive hashing
Random-hyperplane locality-sensitive hashing2
Random-hyperplane locality-sensitive hashingSimHash
Cross-polytope locality-sensitive hashingMinHashp-stable locality-sensitive hashing
Locality-sensitive hashing search2
Locality-sensitive hashing searchMulti-probe LSH search
Data-dependent hashing search
Partition-based approximate neighbor search10
Partition-based approximate neighbor search
Inverted-file vector search5
Inverted-file vector search
IVF-PQ search2
IVF-PQ searchIVF-PQR search
IVF-Flat searchIVF-SQ search
Hierarchical k-means searchInverted multi-index searchSOAR indexingSPANN disk search
Pivot-based approximate neighbor search2
Pivot-based approximate neighbor searchNeighborhood approximation search
Quantization-based approximate neighbor search15
Quantization-based approximate neighbor search
Vector quantization for retrieval8
Vector quantization for retrieval
Additive quantization3
Additive quantizationLocal-search quantizationResidual vector quantization
Product quantization2
Product quantizationOptimized product quantization
Anisotropic vector quantizationScalar vector quantization
Asymmetric distance computationFlat product-quantized searchHNSW-PQ searchHNSW-SQ searchIVF-PQ searchIVF-SQ search
Tree-based approximate neighbor search6
Tree-based approximate neighbor searchApproximate k-d tree searchApproximate vantage-point tree searchHierarchical k-means searchRandom-projection forest searchRandomized k-d tree forest search
Approximate neighbor recall loss2
Approximate neighbor recall lossFiltered retrieval shortfall
Exact nearest-neighbor search4
Exact nearest-neighbor searchExact ball-tree searchExact k-d tree searchExhaustive vector scan
Vector search library10
Vector search libraryAnnoycuVSDiskANN libraryE2LSHFaissFLANNhnswlibNMSLIBScaNN
Benchmark vector retrievalConstruct a nearest-neighbor indexMaintain a nearest-neighbor indexSelect and justify a vector indexSimilarity metric mismatchStale vector indexVector index
Security, privacy and governance106
Security, privacy and governance
AI and data law49
AI and data law
Automated personal-data decision rights3
Automated personal-data decision rightsBrazilian automated-decision review rightsSouth Korean automated-decision rights
Court or regulator decision12
Court or regulator decisionAlice Corp. v. CLS Bank InternationalAustralian Clearview privacy determinationCommission v. Bavarian LagerGoogle v. Oracle API reuse decisionhiQ v. LinkedIn public-access opinionLloyd v. GoogleRyanair v. PR Aviation database rulingSCHUFA credit-scoring rulingThaler v. Comptroller-General of PatentsThaler v. PerlmutterThomson Reuters v. ROSS training-data appeal
Governmental law for ML data flows19
Governmental law for ML data flows
Brazilian personal-data law for ML2
Brazilian personal-data law for MLBrazilian automated-decision review rights
China data and algorithm law for ML2
China data and algorithm law for MLChinese personal-information processing rules
EU text and data mining for ML2
EU text and data mining for MLEU text and data mining exceptions
South Korean privacy and AI law for ML2
South Korean privacy and AI law for MLSouth Korean automated-decision rights
Australian privacy law for MLCanadian privacy law for MLEU AI Act staged applicabilityEU AI governance roles and applicabilityGDPR processing for MLIndian digital personal-data law for MLJapanese privacy law for MLSingapore personal-data law for MLUK data protection for MLUS privacy and sector law for ML
Intellectual property law16
Intellectual property law
Copyright exception8
Copyright exceptionCanadian research and fair dealingEU text and data mining exceptionsGerman text and data mining provisionsIndian fair dealing orientationJapanese non-enjoyment analysis provisionUK non-commercial research analysis exceptionUS fair use
Software license2
Software licenseCopyleft license
Dataset licenseEU text and data mining for MLPatent subject-matter eligibilityTrade-secret protectionTraining-data copyright issues
Biometric-data processing rightsPublic-web collection rights
Data governance4
Data governanceData contractData lineageDataset card
Information security27
Information security
Access control2
Access controlAuthorization
Machine-learning security17
Machine-learning security
Adversarial attack6
Adversarial attack
Model evasion2
Model evasionCarlini–Wagner attack
Data poisoningMembership inferenceModel extraction
AI red teaming2
AI red teamingLLM application red teaming
Adversarial exampleAdversarial robustness evaluationAdversarial trainingJailbreaksLLM guardrailsPrompt injectionPrompt-injection defenseThreat-model an ML workflow
Apache RangerAuthenticationDefense in depthEncryptionLeast privilegeSecret managementThreat modeling
Privacy engineering9
Privacy engineering
Differential privacy3
Differential privacyPrivacy compositionSensitivity-calibrated noise
Data minimizationInvestigate privacy trade-offsPrivacy reviewPrivate aggregation of teacher ensemblesSecure aggregation
Responsible AI16
Responsible AI
Algorithmic fairness4
Algorithmic fairnessEquality of opportunity in classificationEqualized oddsFairness trade-offs
Interpretability and explainability5
Interpretability and explainabilityIntegrated gradientsLIME (local interpretable model-agnostic explanations)Mechanistic interpretabilitySHAP (Shapley additive explanations)
AccountabilityContestabilityDemographic parityHuman oversightModel cardModel risk
Software engineering174
Software engineering
AI-assisted development2
AI-assisted developmentReview AI-generated code
API specification6
API specification
Structured decision interface4
Structured decision interfaceJev Choice primitiveJev Noul primitiveJev Score primitive
DSPy signature
Continuous delivery4
Continuous deliveryCanary releaseRollbackShadow deployment
Developer tool16
Developer tool
Git10
GitGit bisectGit cherry-pickGit fetchGit mergeGit pushGit rebaseGit resetGit revertGit stash
Apache AntApache JMeterApache MavenApache NetBeanspytest
Open-source software13
Open-source softwareANN-BenchmarksApache AmbariApache AtlasApache BigtopApache DataSketchesApache JenaApache LivyApache OpenOfficeApache SupersetApache TinkerPopApache ZeppelinSciPy
Product documentation2
Product documentationPyTorch data-loading documentation
Programming language10
Programming language
CUDA3
CUDACUDA GraphsCUTLASS
Apache GroovyC++JavaTriton kernel languageTypeScriptUnix shell
Python programming5
Python programmingCPython global interpreter lockPython abstract syntax treesPython standard libraryPython warning handling
Retrospective2
RetrospectiveSprint retrospective
Scrum16
Scrum
Product backlog2
Product backlogProduct goal
Sprint5
SprintDaily ScrumSprint planningSprint retrospectiveSprint review
Sprint backlog2
Sprint backlogSprint goal
Definition of DoneIncrementProduct backlog refinementProduct OwnerScrum DevelopersScrum Master
Software frameworks48
Software frameworks
AutoML toolkit7
AutoML toolkitAuto-PyTorchauto-sklearnAutoGluonFLAMLH2O AutoMLTPOT
Bayesian modeling tool5
Bayesian modeling toolArviZPyMCPyStanStan
Deep learning framework13
Deep learning framework
PyTorch5
PyTorchBatch sampler (PyTorch)Data loader (PyTorch)Distributed sampler (PyTorch)Index sampler (PyTorch)
Apache MXNetFlaxJAXKerasMLXPyTorch LightningTensorFlow
Gradient boosting library4
Gradient boosting libraryCatBoostLightGBMXGBoost
Hyperparameter optimization tool6
Hyperparameter optimization toolHyperoptKubeflow KatibOptunaRay TuneSMAC
LLM application framework6
LLM application frameworkDSPyHaystackLangChainLangGraphLlamaIndex
Ray3
RayRay ServeRay Tune
Apache CordovaApache DubbogRPCROS 2
Software library17
Software library
Machine learning library13
Machine learning libraryApache MahoutApache OpenNLPApache SystemDSDeep Graph LibraryGymnasiumHugging Face PEFTHugging Face TransformersPyTorch Geometricscikit-learnSentence TransformersstatsmodelsTokenizers library
Apache CommonsApache HttpComponentsApache Log4j
Software testing9
Software testingDifferential testingFailure injectionFuzz testingIntegration testingMetamorphic testingProperty-based testingStress testingUnit testing
Version control17
Version control
Git repository11
Git repositoryGit branchGit commitGit detached HEADGit ignore rulesGit indexGit reflogGit remoteGit submoduleGit tagGit working tree
Data versioningGit merge conflictGit patchPull requestResolve a version-control conflict
API design1
API design
Dependency management3
Dependency managementReproducible environmentSoftware supply-chain evidence
API designAssess framework and tool fitCode reviewContinuous integrationDependency managementReproducible environmentSoftware supply-chain evidence
System design14
System design
Caching7
Caching
Key-value caching4
Key-value cachingKV cache quantizationPagedAttentionPrefix caching
Application cachingSemantic caching
Architecture decision recordAutoscalingCapacity estimationDesign documentLoad balancingSystem design review
  1. Concepts
  2. Keeping up with ML innovation
  3. Foundations for ML
  4. Mathematics
  5. Optimization
  6. Quasi-Newton method

BFGS

Method

A quasi-Newton update of a Hessian or inverse-Hessian approximation using successive step and gradient differences.

Discuss

Share this page

Read

Docs

  • SciPy minimize: methods and optimization contracts (opens in a new tab)SciPy project
    About this resource

    Study BFGS at notes: BFGS, L-BFGS-B, Newton-CG and trust-region methods.

    Where to read: Notes: BFGS, L-BFGS-B, Newton-CG and trust-region methods

    Checked 2026-10-08

    Authority for this purpose: Maintainer documentation describes the available algorithms, convergence options and implementation-specific contracts.

    Destination inspection: Bounded GET; public DNS/IP pinned; redirect-by-redirect validation; destination text inspection required; reviewed destination title, provenance and concept-specific subject

    Useful destination; no explanatory claim is asserted from this link.

    • Instructional treatment; no claim of historical invention or independent validation of every implementation.

    Reviewed authoritative resource

    Where to read: Notes: BFGS, L-BFGS-B, Newton-CG and trust-region methods

    A first-party destination supports its documented scope; it does not establish independent performance validation or historical priority.

Related concepts

Broader

  • Quasi-Newton methodMethod

Related

  • SciPyTool
  • L-BFGS-BMethod
MLE Path moduleBFGS · 1 lesson
  • Audit curvature updates, box-bound stationarity and trust-region agreement
See the lessons »

Books to go deeper

Book

Deep Learning

Ian Goodfellow, Yoshua Bengio and Aaron Courville · 2016 edition

Mathematical and neural-network foundations. Predates modern Transformers.

View on Amazon (affiliate link) ↗
Book

Probabilistic Machine Learning: An Introduction

Kevin P. Murphy · 2022 edition

Probability, inference and predictive modeling with uncertainty.

View on Amazon (affiliate link) ↗

As an Amazon Associate I earn from qualifying purchases.

Loading discussion…

MLE Path

Learn ML concepts, prepare for ML interviews and find ML roles.

Interview prepOverviewYour planQuestionsCompaniesMocks
ConceptsConcept mapLearning pathsModulesTopic maps
JobsJob boardResume scanEmployers
MLE PathWhat's newPricingAboutAffiliateYour account
PrivacyTermsRefundsFAQContact