From local explanations to global understanding with explainable AI for trees

2020/01/17 by Scott M. Lundberg, Scott Lundberg, Gabriel Erion +11 · 9,466 citations
Computer Science · #Bayesian Modeling and Causal Inference #Computer science #Explainable Artificial Intelligence (XAI)

paper · open access · doi:10.1038/s42256-019-0138-9

published in Nature Machine Intelligence 2(1), 56-67 (Nature Portfolio)

crossref issued 2020/01/17 · crossref published 2020/01/17 · crossref published-online 2020/01/17 · openalex publication_date 2020/01/17 · crossref created 2020/01/17 · crossref deposited 2022/12/06 · openalex created_date 2025/10/10 · crossref indexed 2026/07/30 · openalex updated_date 2026/08/06

Citations

• Understanding Neural Networks Through Deep Visualization
• Monotonic solutions of cooperative games
• Explaining prediction models and individual predictions with feature contributions
• How to Explain Individual Classification Decisions
• Heart Disease and Stroke Statistics—2016 Update
• Model Agnostic Supervised Local Explanations
• On a Method to Measure Supervised Multiclass Model’s Interpretability: Application to Degradation Diagnosis (Short Paper)
• A Benchmark for Interpretability Methods in Deep Neural Networks

Cited by

• The Future of Sensitivity Analysis: An essential discipline for systems modeling and policy support
• An Explainable Deep Learning Model for Focal Liver Lesion Diagnosis Using Multiparametric MRI
• Global geometry of chemical graph neural network representations in terms of chemical moieties
• Machine Learning Predicts Hepatocellular Carcinoma Risk from Routine Clinical Data: A Large Population-Based Multicentric Study
• Explaining the data or explaining a model? Shapley values that uncover non-linear dependencies
• Modern machine‐learning for binding affinity estimation of protein–ligand complexes: Progress, opportunities, and challenges
• The timescale of nematode evolution reveals late and convergent radiations of parasitism
• Global density and biomass of arbuscular mycorrhizal fungal networks
• Predictors of loneliness at school: An explainable artificial intelligence approach on a large-scale cross-cultural assessment
• Delivering equity in low-carbon multisector infrastructure planning
• Machine learning-enhanced high-resolution exposure assessment of ultrafine particles
• Barlow Twins deep neural network for advanced 1D drug–target interaction prediction
• In situ molecular profiles of glomerular cells by integrated imaging mass spectrometry and multiplexed immunofluorescence microscopy
• Travel time prediction from sparse open data
• ShaplEIG: Bayesian Experimental Design for Shapley Value Estimation
• Compound soil and atmospheric drought (CSAD) events and CO 2 fluxes of a mixed deciduous forest: the occurrence, impact, and temporal contribution of main drivers
• Human-Centered Explainable AI (XAI): From Algorithms to User Experiences
• Platform-induced time-space trade-offs in ride-hailing: Multi-homing as a response to operational constraints
• OptiLIME: Optimized LIME Explanations for Diagnostic Computer Algorithms
• China’s 1 km Daily Surface Soil Moisture Fusion Dataset (2000–2025) Based on Explainable Machine Learning
• A multi‐source remote sensing approach to identify and predict delayed succession in human‐dominated tropical landscapes
• Understanding Heart-Failure Patients EHR Clinical Features via SHAP Interpretation of Tree-Based Machine Learning Model Predictions
• Towards actionable hypotension prediction -- predicting catecholamine therapy initiation in the intensive care unit
• Tree Ensemble Explainability through the Hoeffding Functional Decomposition and TreeHFD Algorithm
• Moment connectedness and driving factors in the energy-food nexus: A time-frequency perspective
• Global high-resolution ultrafine particle number concentrations through data fusion with machine learning
• Disentangled Explanations of Neural Network Predictions by Finding Relevant Subspaces
• Global hotspots of mycorrhizal fungal richness are poorly protected
• AI-powered dissecting of carbohydrate isomers by a flexoelectric gated nanopore tweezer
• Enhanced forecasting of bird nocturnal migration intensity in relation to previous days and synoptic weather patterns
• GroupSHAP-Guided Integration of Financial News Keywords and Technical Indicators for Stock Price Prediction
• Explaining by Removing: A Unified Framework for Model Explanation
• SHAP Meets Tensor Networks: Provably Tractable Explanations with Parallelism
• Competing narratives in AI ethics: a defense of sociotechnical pragmatism
• No Intelligence Without Statistics: The Invisible Backbone of Artificial Intelligence
• EEG-based brain age prediction in infants–toddlers: Implications for early detection of neurodevelopmental disorders
• Deep Learning-Based Extraction of Promising Material Groups and Common Features from High-Dimensional Data: A Case of Optical Spectra of Inorganic Crystals
• A tutorial on discovering and quantifying the effect of latent causal sources of multimodal EHR data
• Phenome-Wide Multi-Omics Integration Uncovers Distinct Archetypes of Human Aging
• Enhanced lithological mapping in arid crystalline regions using explainable AI and multi-spectral remote sensing data
• Interpretable Machine Learning for Cognitive Aging: Handling Missing Data and Uncovering Social Determinant
• Machine learning for anxiety and depression profiling and risk assessment in the aftermath of an emergency
• SKALE: An Interpretable Multiscale Machine Learning Model for Decoding Phase‐Specific Protein Aggregation in Neurodegenerative Proteinopathies
• Faithful and Interpretable Explanations for Complex Ensemble Time Series Forecasts using Surrogate Models and Forecastability Analysis
• Increasing river sediment concentration and flux across the pan-Arctic
• Host traits and environmental factors shape infection heterogeneity in wild rat–protozoa networks
• Can Machine Learning Target Health Care Fraud? Evidence From Medicare Hospitalizations
• TreeNet: Layered Decision Ensembles
• On Predictive Explanation of Data Anomalies
• Elucidating the Grey Atmosphere: SHAP Value Analysis of a Random Forest Atmospheric Neutral Density Model
• CardioForest: An Explainable Ensemble Learning Model for Automatic Wide QRS Complex Tachycardia Diagnosis from ECG
• Private and public school efficiency gaps in Latin America-A combined DEA and machine learning approach based on PISA 2022
• Determinants of Latin American students academic resilience-Insights based on PISA 2022 using an explainable machine learning approach
• A lipid atlas of the human kidney
• Fast and accurate explanations of distance-based classifiers by uncovering latent explanatory structures
• Transparency of medical artificial intelligence systems
• Model-agnostic interpretation by visualization of feature perturbations
• Data-driven composition-only machine learning for high-performance solid-state electrolytes
• A predictive model for household displacement duration after disasters
• Bridging human and machine intelligence: Reverse-engineering radiologist intentions for clinical trust and adoption
• Vocal repertoire and individuality in the plains zebra ( Equus quagga )
• Optimizing urban park locations with addressing environmental justice in park access and utilization by using dynamic demographic features derived from mobile phone data
• A Quantum-Chemical Bonding Database for Solid-State Materials
• The Data Gradient: Quantifying Dataset Influence in End-to-End Machine Learning Systems
• Precipitation threshold-driven shifts in dominant controls of ecosystem nitrogen retention
• Understanding the Drivers of Drought Onset and Intensification in the Canadian Prairies: Insights from Explainable Artificial Intelligence (XAI)
• Machine Learning-based beam delivery time model for Mevion 250i with Hyperscan technology
• Efficient nonparametric statistical inference on population feature importance using Shapley values
• 'Entanglement' -- A new dynamic metric to measure team flow
• Explainable integration of process and optical parameters for plasma etch depth prediction
• Phenology-Aligned multi-task temporal fusion framework for satellite-based triple-seasonal rice yield estimation in Southeast Asia
• Photosynthetic optimum temperature plays a minor role in the increase of terrestrial carbon uptake from 2000 to 2019
• High-throughput optimized prime editing mediated endogenous protein tagging for pooled imaging of protein localization
• Noncanonical genetic markers resolve the pre-GOE emergence of aerobic bacteria in Earth’s history
• Development and validation of machine learning models to predict vancomycin- and teicoplanin-associated acute kidney injury: A retrospective, multicenter study
• Presurgical structural connectivity predicts postsurgical cognitive impairment in glioma patients
• Unifying local and global model explanations by functional decomposition of low dimensional structures
• Development of a natural language processing algorithm to extract social determinants of health from clinician notes
• BacTermFinder: a comprehensive and general bacterial terminator finder using a CNN ensemble
• Unraveling the drivers of leptospirosis risk in Thailand using machine learning
• Satellite-based evidence of recent decline in global forest recovery rate from tree mortality events
• A systematic exploration of bacterial form I rubisco maximal carboxylation rates
• Expanding the flexibility of base editing for high-throughput genetic screens in bacteria
• Improved prediction of bacterial CRISPRi guide efficiency from depletion screens through mixed-effect machine learning and data integration
• Rapid changes in global river particulate organic carbon flux
• Feature selection with a lexicographic social ranking method
• Thriving Areas in Temperate Coastal Systems: Novel Insights for Marine Conservation
• Fréchet Geodesic Boosting
• Predicting Chest Radiograph Findings from Electrocardiograms Using Interpretable Machine Learning
• Cognitive Linguistic Identity Fusion Score (CLIFS): A Scalable Cognition-Informed Approach to Quantifying Identity Fusion from Text
• Enhancing Antimicrobial Molecule Prediction via Dynamic Routing Capsule Networks and Multi-Source Molecular Embeddings
• Uncovering multi-technology convergence patterns with hypergraphs: Evolution and prediction using patent data
• Top-k Feature Importance Ranking
• Exploring the Relationship between Brain Hemisphere States and Frequency Bands through Deep Learning Optimization Techniques
• Tractability Frontiers of the Shapley Value for Aggregate Conjunctive Queries
• Sampling Permutations for Shapley Value Estimation
• TwoWin-SOVA: Two Windows Discrete Cosine Transform and Synthetic Minority Oversampling Technique One-Versus-All Ensemble Classifiers for Imbalanced Hyperspectral Image Explainable Classification
• Prediction of post-stroke urinary tract infection risk in immobile patients using machine learning: an observational cohort study
• Neurosymbolic AI Transfer Learning Improves Network Intrusion Detection
• MetaLLMix : An XAI Aided LLM-Meta-learning Based Approach for Hyper-parameters Optimization
• Uncertainty Awareness and Trust in Explainable AI- On Trust Calibration using Local and Global Explanations
• NBDT: Neural-Backed Decision Trees
• Assessing equity in infrastructure investment distribution among U.S. cities
• Causal Beam Selection for Reliable Initial Access in AI-driven Beam Management
• Recurrent visitations reveal selectivity beyond the 15-Minute City vision
• Causal SHAP: Feature Attribution with Dependency Awareness through Causal Discovery
• XAI-Driven Machine Learning System for Driving Style Recognition and Personalized Recommendations
• Shapley Flow: A Graph-based Approach to Interpreting Model Predictions
• Delta-Audit: Explaining What Changes When Models Change
• Enhancing XAI Interpretation through a Reverse Mapping from Insights to Visualizations
• microRNA‐based predictor for diagnosis of frontotemporal dementia
• Machine Learning-Guided Design of Rhenium Tricarbonyl Complexes for Next-Generation Antibiotics
• EEG-MSAF: An Interpretable Microstate Framework uncovers Default-Mode Decoherence in Early Neurodegeneration
• Shapley Values: Paired-Sampling Approximations
• UbiQTree: Uncertainty Quantification in XAI with Tree Ensembles
• Understanding Global Feature Contributions With Additive Importance Measures
• Integrating attention into explanation frameworks for language and vision transformers
• Attribution Explanations for Deep Neural Networks: A Theoretical Perspective
• Study of Robust Features in Formulating Guidance for Heuristic Algorithms for Solving the Vehicle Routing Problem
• FIRE: Faithful Interpretable Recommendation Explanations
• Hybrid quantum tensor networks for aeroelastic applications
• Unveiling Location-Specific Price Drivers: A Two-Stage Cluster Analysis for Interpretable House Price Predictions
• Improving Hospital Risk Prediction with Knowledge-Augmented Multimodal EHR Modeling
• Explainable Machine Learning in Credit Risk Management
• Forest-Guided Clustering -- Shedding Light into the Random Forest Black Box
• Mangrove sediment carbon burial offset by methane emissions from mangrove tree stems
• Safer and Smarter: Leveraging Interpretation-Guided Modeling and Data Merging of Disease and Environmental Data for Plant Disease Risk Prediction
• Model-Agnostic Explanations using Minimal Forcing Subsets
• Interpretable QSAR, External PubChem Validation, and Coordination-Aware Docking Enable Tiered Prioritization of Carbonic Anhydrase I Inhibitors
• Do not explain without context: addressing the blind spot of model explanations
• A geological timescale for bacterial evolution and oxygen adaptation
• Improved Feature Importance Computations for Tree Models: Shapley vs. Banzhaf
• On the Importance of Domain-specific Explanations in AI-based Cybersecurity Systems (Technical Report)
• From predictions to prescriptions: A data-driven response to COVID-19
• Lateral shifting behavior of vehicles at horizontal curves and its influencing factors: application of LightGBM and SHAP
• FakeWake: Understanding and Mitigating Fake Wake-up Words of Voice Assistants
• StylOch at PAN: Gradient-Boosted Trees with Frequency-Based Stylometric Features
• Machine learning in Alzheimer’s disease genetics
• Temporal quality degradation in AI models
• Local MDI+: Local Feature Importances for Tree-Based Models
• Distinguishing Orbiting and Infalling Dark Matter Particles with Machine Learning
• TaylorPODA: A Taylor Expansion-Based Method to Improve Post-Hoc Attributions for Opaque Models
• Kinetically accessible 1D magnetic chains of transition-metal chalcogenides and halides on van der Waals surfaces
• Digital Twin-Assisted Explainable AI for Robust Beam Prediction in mmWave MIMO Systems
• A Taxonomy for Design and Evaluation of Prompt-Based Natural Language Explanations
• Entity-Specific Cyber Risk Assessment using InsurTech Empowered Risk Factors
• Understanding the diversity of the metal-organic framework ecosystem
• Using Eye-tracking Data to Predict Situation Awareness in Real Time during Takeover Transitions in Conditionally Automated Driving
• Look behind the Censorship: Reposting-User Characterization and Muted-Topic Restoration
• From Street Form to Spatial Justice: Explaining Urban Exercise Inequality via a Triadic SHAP-Informed Framework
• Beyond Charging Anxiety: An Explainable Approach to Understanding User Preferences of EV Charging Stations Using Review Data
• TILES-2018 Sleep Benchmark Dataset: A Longitudinal Wearable Sleep Data Set of Hospital Workers for Modeling and Understanding Sleep Behaviors
• Neuron Shapley: Discovering the Responsible Neurons
• Improving Early Sepsis Prediction with Multi Modal Learning
• Human-Centered Explainability in Interactive Information Systems: A Survey
• Shaping Scientific Explanations to Expert Perspectives with Persona-Conditioned Reinforcement Learning
• Stylometry recognizes human and LLM-generated texts in short samples
• Towards transparent and data-driven fault detection in manufacturing: A case study on univariate, discrete time series
• Data Science Principles for Interpretable and Explainable AI
• Why Uncertainty Calibration Matters for Reliable Perturbation-based Explanations
• The Most Important Features in Generalized Additive Models Might Be Groups of Features
• Pattern-Based Graph Classification: Comparison of Quality Measures and Importance of Preprocessing
• How about electric vehicle? Sensing owners’ experiences and attitudes through online short video
• Towards Desiderata-Driven Design of Visual Counterfactual Explainers
• Towards real-time additive-free dopamine detection at 10^-8 mM with hardware accelerated platform integrated on camera
• Feature Removal Is a Unifying Principle for Model Explanation Methods
• Machine learning models reveal environmental and genetic factors associated with the plant circadian clock
• Regression-adjusted Monte Carlo Estimators for Shapley Values and Probabilistic Values
• Interpretation of machine learning models using shapley values: application to compound potency and multi-target activity predictions
• Accelerated design of nickel-cobalt based catalysts for CO2 hydrogenation with human-in-the-loop active machine learning
• Systematic exploration of bacterial form I rubisco maximal carboxylation rates
• An Interpretable Machine Learning Approach in Predicting Inflation Using Payments System Data: A Case Study of Indonesia
• RETENTION: Resource-Efficient Tree-Based Ensemble Model Acceleration with Content-Addressable Memory
• Applying XAI based unsupervised knowledge discovering for Operation modes in a WWTP. A real case: AQUAVALL WWTP
• Why is My Route Different Today? An Algorithm for Explaining Route Selection
• A Unified Framework for Provably Efficient Algorithms to Estimate Shapley Values
• CoFrNets: Interpretable Neural Architecture Inspired by Continued Fractions
• MVP-Shapley: Feature-based Modeling for Evaluating the Most Valuable Player in Basketball
• Machine Assistance for Credit Card Approval? Random Wheel can Recommend and Explain
• The Shapley Value of coalition of variables provides better explanations
• Reasoning about Counterfactuals and Explanations: Problems, Results and Directions
• Prediction via Shapley Value Regression
• Spatial-temporal insights into gender gaps in East Asian ride-hailing: Workload, efficiency, nighttime safety, and operational patterns
• Inferring feature importance with uncertainties in high-dimensional data
• DumbleDR: Predicting User Preferences of Dimensionality Reduction Projection Quality
• Neurosymbolic Artificial Intelligence for Robust Network Intrusion Detection: From Scratch to Transfer Learning
• Shapley variable importance clouds for interpretable machine learning
• ShaTS: A Shapley-based Explainability Method for Time Series Artificial Intelligence Models applied to Anomaly Detection in Industrial Internet of Things
• Shapley explainability on the data manifold
• Understanding the European energy crisis through structural causal models
• Machine Learning Framework for Characterizing Processing-Structure Relationship in Block Copolymer Thin Films
• Data Model Design for Explainable Machine Learning-based Electricity Applications
• Design Rules for Optimizing Quaternary Mixed-Metal Chalcohalides
• Amortized Linear-time Exact Shapley Value for Product-Kernel Methods
• Children's Mental Models of AI Reasoning: Implications for AI Literacy Education
• Unveil Sources of Uncertainty: Feature Contribution to Conformal Prediction Intervals
• Integrating molecular descriptors for enhanced prediction: Shedding light on the potential of pH to model hydrated electron reaction rates for organic compounds
• Machine-Learning-Based Prediction of Plant Cuticle–Air Partition Coefficients for Organic Pollutants: Revealing Mechanisms from a Molecular Structure Perspective
• Data Mining and Computational Screening of Rashba-Dresselhaus Splitting and Optoelectronic Properties in Two-Dimensional Perovskite Materials
• Surrogate Interpretable Graph for Random Decision Forests
• Particle identification in the GlueX detector with machine learning
• Enhancing operations management through smart sensors: measuring and improving well-being, interaction and performance of logistics workers
• Systematic interrogation of CRISPR antimicrobials in Klebsiella pneumoniae reveals nuclease-, guide- and strain-dependent features influencing antimicrobial activity
• Can Global XAI Methods Reveal Injected Behaviours in LLMs? SHAP vs Rule Extraction vs RuleSHAP
• Targeted design of synthetic enhancers for selected tissues in the Drosophila embryo
• The identity of social impact venture capitalists: exploring social linguistic positioning and linguistic distinctiveness through text mining
• SIM-Shapley: A Stable and Computationally Efficient Approach to Shapley Value Approximation
• Unwrapping The Black Box of Deep ReLU Networks: Interpretability, Diagnostics, and Simplification
• Local Shapley: Model-Induced Locality and Optimal Reuse in Data Valuation
• The role of artificial intelligence in business transformation: A case of pharmaceutical companies
• Overview and practical recommendations on using Shapley Values for identifying predictive biomarkers via CATE modeling
• Dynamic Forecasting and Temporal Feature Evolution of Stock Repurchases in Listed Companies Using Attention-Based Deep Temporal Networks
• comprisk: A scikit-learn-compatible Python toolkit for competing-risks survival analysis
• Global burden and country-level predictive correlates of spinal cord injuries attributable to transport injuries among youths and young adults, 1990 to 2021: A Global Burden of Disease 2021 analysis
• Modeling Dispositional and Initial learned Trust in Automated Vehicles with Predictability and Explainability
• Interpreting Contrastive Embeddings in Specific Domains with Fuzzy Rules
• Knowledge graph construction and application in geosciences: A review
• Peak-Based Nuclide Identification in HPGe γ-Spectrometry with Machine Learning and SHAP
• Who engages? Machine learning insights into digital mindfulness-based intervention for generalized anxiety disorder
• A SHAP-enhanced XGBoost model for interpretable prediction of coseismic landslides
• Unbox the Black-box for the Medical Explainable AI via Multi-modal and Multi-centre Data Fusion: A Mini-Review, Two Showcases and Beyond
• Better Model Selection with a new Definition of Feature Importance
• modelimportance: An R package for evaluating model importance within a multi-model ensemble
• Emotion Recognition in Contemporary Dance Performances Using Laban Movement Analysis
• Formal and Empirical Study of Metadata-Based Profiling for Resource Management in the Computing Continuum
• Explanation format does not matter; but explanations do -- An Eggsbert study on explaining Bayesian Optimisation tasks
• Explanations Go Linear: Post-hoc Explainability for Tabular Data with Interpretable Meta-Encoding
• Dance Style Recognition Using Laban Movement Analysis
• AI-Derived Reproductive Phenotypes and Explainable ML for Concurrent Early Multimorbidity in U.S. Women: NHANES 2017-March 2020
• A governance model for the application of AI in health care
• ConfoundingSHAP: Quantifying confounding strength in causal inference
• Explainable Patterns in Cryptocurrency Microstructure
• Interpretable XGBoost-based predictions of shear wave velocity from CPTu data
• DiCE-Extended: A Robust Approach to Counterfactual Explanations in Machine Learning
• Natural Language Embeddings of Synthesis and Testing conditions Enhance Glass Dissolution Prediction
• Learning to fuse: dynamic integration of multi-source data for accurate battery lifespan prediction
• Explainable AI for UAV Mobility Management: A Deep Q-Network Approach for Handover Minimization
• Quantification of the influence of laser shock peening parameters on residual stresses in thick AA 7050 specimens: An experimental and explainable machine learning-based approach
• Identifying and prioritizing potential human-infecting viruses from their genome sequences
• Systemic Flakiness: An Empirical Analysis of Co-Occurring Flaky Test Failures
• iAFPs-EnC-GA: Identifying antifungal peptides using sequential and evolutionary descriptors based multi-information fusion and ensemble learning approach
• Neural Forecasting of the Italian Sovereign Bond Market with Economic News
• mirkwood: Fast and Accurate SED Modeling Using Machine Learning
• Enhancing interpretability in neural networks for nuclear power plant fault diagnosis: A comprehensive analysis and improvement approach
• Cryptogenic stroke and migraine: using probabilistic independence and machine learning to uncover latent sources of disease from the electronic health record
• Reliable Classification with Conformal Learning and Interval-Type 2 Fuzzy Sets
• Machine learning land surface temperature downscaling method based on Landsat 9 and Sentinel-2 satellite feature interaction
• Quantifying drivers and compound effects of multivariate drought heat events in global river basins
• Interpreted machine learning in fluid dynamics: explaining relaminarisation events in wall-bounded shear flows
• Deep Neural Networks and Tabular Data: A Survey
• Eco-efficiency as a Catalyst for Citizen Co-production: Evidence from Chinese Cities
• Enhancing Explainability and Reliable Decision-Making in Particle Swarm Optimization through Communication Topologies
• A Taxonomy of Questions for Critical Reflection in Machine-Assisted Decision-Making
• Quantum Phases Classification Using Quantum Machine Learning with SHAP-Driven Feature Selection
• Unequal futures: Projecting lake trophic states in China under landscape transformation scenarios