Making the Black Box More Transparent: Understanding the Physical Implications of Machine Learning

2019/08/22 by Amy McGovern, Ryan Lagerquist, David John Gagne +5 · 560 citations
Computer Science · Earth and Planetary Sciences · Environmental Science · #Artificial intelligence #Black box #Class (philosophy) #Climate variability and models #Computer science #Data Analysis with R #Machine learning #Meteorological Phenomena and Simulations #Novelty #Toolbox #Variety (cybernetics) #Visualization

paper · pdf · doi:10.1175/bams-d-18-0195.1

published in Bulletin of the American Meteorological Society 100(11), 2175-2199 (American Meteorological Society)

openalex publication_date 2019/08/22 · openalex created_date 2019/08/29 · openalex updated_date 2026/08/05

Abstract

Abstract This paper synthesizes multiple methods for machine learning (ML) model interpretation and visualization (MIV) focusing on meteorological applications. ML has recently exploded in popularity in many fields, including meteorology. Although ML has been successful in meteorology, it has not been as widely accepted, primarily due to the perception that ML models are “black boxes,” meaning the ML methods are thought to take inputs and provide outputs but not to yield physically interpretable information to the user. This paper introduces and demonstrates multiple MIV techniques for both traditional ML and deep learning, to enable meteorologists to understand what ML models have learned. We discuss permutation-based predictor importance, forward and backward selection, saliency maps, class-activation maps, backward optimization, and novelty detection. We apply these methods at multiple spatiotemporal scales to tornado, hail, winter precipitation type, and convective-storm mode. By analyzing such a wide variety of applications, we intend for this work to demystify the black box of ML, offer insight in applying MIV techniques, and serve as a MIV toolbox for meteorologists and other physical scientists.

Cited by

• Machine learning methods for postprocessing ensemble forecasts of wind gusts: A systematic comparison
• Predicting Cloud‐To‐Ground Lightning in the Western United States From the Large‐Scale Environment Using Explainable Neural Networks
• Exploring a Data‐Driven Approach to Identify Regions of Change Associated With Future Climate Scenarios
• A shifting climate: New paradigms and challenges for (early career) scientists in extreme weather research
• Can Machines Learn How Clouds Work? The Epistemic Implications of Machine Learning Methods in Climate Science
• Interpretable Machine Learning-Derived Spectral Indices for Vegetation Monitoring
• Winter Precipitation Type Diagnosis and Uncertainty Quantification with a Physically Consistent Machine Learning Method
• Deep Learning Emulation of Multivariate Climate Indices: A Case Study of the Fire Weather Index in the Iberian Peninsula
• The Benchmarking Epistemology: Construct Validity for Evaluating Machine Learning Models
• Explaining heatwaves with machine learning
• Fire Intensity and spRead forecAst (FIRA): A Machine Learning Based Fire Spread Prediction Model for Air Quality Forecasting Application
• Synergistic impact of marine heat waves and rapid intensification exacerbates tropical cyclone destructive power worldwide
• Physics-informed machine learning: case studies for weather and climate modelling
• A review of radar-based nowcasting of precipitation and applicable machine learning techniques
• Trust and trustworthy artificial intelligence: A research agenda for AI in the environmental sciences
• Transparent Machine Learning: Training and Refining an Explainable Boosting Machine to Identify Overshooting Tops in Satellite Imagery
• FrontFinder AI: Efficient Identification of Frontal Boundaries over the Continental United States and NOAA’s Unified Surface Analysis Domain Using the UNET3+ Model Architecture
• Analyzing the effects of data splitting and covariate shift on machine learning based streamflow prediction in ungauged basins
• WeatherBench: A benchmark dataset for data-driven weather forecasting
• Exploiting deep learning in forecasting the occurrence of severe haze in Southeast Asia
• A review of machine learning applications in wildfire science and management
• Indicator Patterns of Forced Change Learned by an Artificial Neural Network
• A mask R-CNN model for reidentifying extratropical cyclones based on quasi-supervised thought. [europepmc]
• Towards implementing artificial intelligence post-processing in weather and climate: proposed actions from the Oxford 2019 workshop. [europepmc]
• Solving musculoskeletal biomechanics with machine learning. [europepmc]
• A fully automated method of human identification based on dental panoramic radiographs using a convolutional neural network. [europepmc]
• COVID-19 mortality risk assessments for individuals with and without diabetes mellitus: Machine learning models integrated with interpretation framework. [europepmc]
• Multi-fidelity information fusion with concatenated neural networks. [europepmc]
• Forecasting large-scale circulation regimes using deformable convolutional neural networks and global spatiotemporal climate data. [europepmc]
• Analyzing the serum of hemodialysis patients with end-stage chronic kidney disease by means of the combination of SERS and machine learning. [europepmc]
• Physics-embedded inverse analysis with algorithmic differentiation for the earth's subsurface. [europepmc]
• Using Machine Learning Algorithms to Pool Data from Meta-Analysis for the Prediction of Countermovement Jump Improvement. [europepmc]
• An automated surgical decision-making framework for partial or radical nephrectomy based on 3D-CT multi-level anatomical features in renal cell carcinoma. [europepmc]
• Explainable sequence-to-sequence GRU neural network for pollution forecasting. [europepmc]
• Water quality prediction based on sparse dataset using enhanced machine learning. [europepmc]
• A machine learning model that outperforms conventional global subseasonal forecast models. [europepmc]
• A machine learning paradigm for necessary observations to reduce uncertainties in aerosol climate forcing. [europepmc]
• Predicting Cloud-To-Ground Lightning in the Western United States From the Large-Scale Environment Using Explainable Neural Networks. [europepmc]
• Fire Intensity and spRead forecAst (FIRA): A Machine Learning Based Fire Spread Prediction Model for Air Quality Forecasting Application. [europepmc]
• Selected Topics in Time Series Forecasting: Statistical Models vs. Machine Learning. [europepmc]
• Bayesian deep learning for probabilistic aquifer vulnerability and uncertainty prediction. [europepmc]
• Bridging computational power and environmental challenges: a perspective on neural network predictive models for environmental engineering. [europepmc]
• Increased hailstorms in cities through cell merger mechanism across North America and East Asia. [europepmc]
• Curtain grouting volume prediction using a Bayesian-optimized stacking ensemble model with SHAP analysis. [europepmc]

Related

• Physics-informed machine learning: case studies for weather and climate modelling
• Interpretable Machine Learning for Weather and Climate Prediction: A Survey
• A Physics-guided Multimodal Transformer Path to Weather and Climate Sciences
• Deep Learning Techniques in Extreme Weather Events: A Review
• AI Foundation Models for Weather and Climate: Applications, Design, and Implementation