MIT Common Ground 6.S891/6.S893/12.S992 AI for Climate Action

Spring 2026

---

Class Schedule

** subject to change **

Last updated: 02/27/2026

Tracks (Weeks 4-12): 12.S992: Climate Models | 6.S891: Biodiversity and Environment | 6.S893 Power & Energy Systems

Jump to:

(merged weeks) Week 1 | Week 2 | Week 3

(forked weeks) Week 4 | Week 5 | Week 6 | Week 7 | Week 8 (Spring Break) | Week 9 | Week 10 | Week 11 | Week 12

(merged weeks) Week 13 | Week 14 | Week 15

Week / Date	Track / Focus	Topic	Instructor	Presented Papers	Optional Readings
Week 1
Mon 02/02	Merged	Intro to AI and Climate Change (adaptation, mitigation) + Logistics Slides Recording	Donti	Syllabus	Tackling Climate Change with Machine Learning Climate Change and AI: Recommendations for Government Action Aligning Artificial Intelligence with Climate Change Mitigation
Wed 02/04	Merged	Application-driven innovation in Machine Learning Slides Recording	Beery		Application-Driven Innovation in Machine Learning Machine Learning that Matters The Values Encoded in Machine Learning Research
Week 2
Mon 02/09	Merged	Mitigation of Climate Change (IPCC WG3) Human impacts and infrastructure Slides Recording	Donti		Summary for Policymakers - Climate Change 2022: Mitigation of Climate Change
Wed 02/11	Merged	Impacts, Adaptation and Vulnerability (IPCC WG2) The natural world Slides Recording	Beery		Summary for Policymakers - Climate Change 2022: Impacts, Adaptation and Vulnerability
Week 3
Tue 02/17 (MIT Monday)	Merged	The Physical Science Basis (IPCC WG1) Data methods Models and observations Slides Recording	Bodner		Summary for Policymakers - Climate Change 2021: The Physical Science Basis
Wed 02/18	Merged	AI powered climate modeling for mitigation and adaptation Slides Recording	Bodner		AI-empowered next-generation multiscale climate modelling for mitigation and adaptation
Week 4 - Forked Lessons (Week 1 of Forked)
Mon 02/23	Climate Models	Introduction to Earth System Models	Bodner		Advances and challenges in climate modeling Evolution and prospects of Earth system models: Challenges and opportunities Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization
	Biodiversity	Introduction to biodiversity, ecosystems, and evolution Slides	Beery	GPAI AI and Biodiversity Report People-centric AI for Conserving Biodiversity Machine Learning for Wildlife Conservation Getting the measure of biodiversity The carbon footprint of AI in Ecology
	Power & Energy	Lecture: Introduction to Power and Energy Systems Slides Recording	Donti	Electricity 101: Terms and Definitions Machine Learning for Sustainable Energy Systems
Wed 02/25	Climate Models	Introduction to Earth System Models Recording	Bodner	Climate-invariant machine learning Rewiring climate modeling with machine learning emulators CERA: A Framework for Improved Generalization of Machine Learning Models to Changed Climates
	Biodiversity	Introduction to biodiversity, ecosystems, and evolution	Beery	Seven biodiversity shortfalls The future of biodiversity Essential Biodiversity Variables
	Power & Energy	Lecture: Situational Awareness (Forecasting, State Estimation, and Predictive Maintenance) Slides Recording	Donti
Week 5 - Forked Lessons (Week 2 of Forked)
Mon 03/02	Climate Models	Future climate scenarios & multi-model ensembles	Bodner		Parametrization in weather and climate models A Framework for Hybrid Physics-AI Coupled Ocean Models
	Biodiversity	Monitoring and Evaluation	Beery	Taxonomic bias in biodiversity data and its implications for conservation WILDS: A Benchmark of In-the-Wild Distribution Shifts The devil is in the tails: Fine-grained classification in the wild Fill-Up: Balancing Long-Tailed Data with Generative Models
	Power & Energy	Paper Presentations: State Estimation, Predictive Maintenance, and Anomaly Detection AI topics: Physics-informed ML, anomaly detection, computer vision	Donti	Physics-guided Deep Learning for Power System State Estimation Predicting battery end of life from solar off-grid system field data using machine learning Machine vision for natural gas methane emissions detection using an infrared camera	Machine Learning for the New York City Power Grid PFΔ: A Benchmark Dataset for Power Flow under Load, Generation, and Topology Variations Non-Intrusive Load Monitoring: A Review
Wed 03/04	Climate Models	Future climate scenarios & multi-model ensembles	Bodner	Generalizable Neural-Network Parameterization of Mesoscale Eddies in Idealized and Global Ocean Models Neural-Network Parameterization of Subgrid Momentum Transport in the Atmosphere Learning Closed-Form Equations for Subgrid-Scale Closures From High-Fidelity Data: Promises and Challenges
	Biodiversity	Monitoring and Evaluation	Beery	Meta-learning to address diverse Earth observation problems across resolutions WildlifeDatasets: An open-source toolkit for animal re-identification Generate, Transduct, Adapt: Iterative Transduction with VLMs
	Power & Energy	Paper Presentations: Short-term Forecasting of Demand and Renewables AI topics: Decision-cognizant learning, probabilistic methods, missing data methods, multi-modal learning, distribution shift, interpretable and explainable ML	Donti	Task-based end-to-end model learning in stochastic optimization Marginalize, Rather than Impute: Probabilistic Wind Power Forecasting with Incomplete Data Sky image-based solar forecasting using deep learning with heterogeneous multi-location data: Dataset fusion versus transfer learning	Short-term Prediction and Filtering of Solar Power Using State-Space Gaussian Processes Explainability and Interpretability in Electric Load Forecasting Using Machine Learning Techniques – A Review
Week 6 - Forked Lessons (Week 3 of Forked)
Mon 03/09	Climate Models	Parameterizing subgrid effects, generative AI, super-resolution	Bodner		Traditional Data Assimilation and Its Effective Integration with Meteorological Deep Learning Machine learning in ocean data assimilation: Advances, gaps and the road to operations
	Biodiversity	Distribution shift and bias	Beery	A deep active learning system for species identification and counting in camera trap images DataS3: Dataset Subset Selection for Specialization Domain-Adversarial Training of Neural Networks
	Power & Energy	Lecture: Power Grid Optimization, Control, and Markets	Donti
Wed 03/11	Climate Models	Parameterizing subgrid effects, generative AI, super-resolution	Bodner	Generative Data Assimilation for Surface Ocean State Estimation From Multi-Modal Satellite Observations Deep Learning of Systematic Sea Ice Model Errors From Data Assimilation Increments Deterministic and stochastic tendency adjustments derived from data assimilation and nudging
	Biodiversity	Distribution shift and bias	Beery	Consensus-Driven Active Model Selection The Search for Squawk: Agile Modeling in Bioacoustics Spatial Implicit Neural Representations for Global-Scale Species Mapping
	Power & Energy	Paper presentations: Fast solutions for optimal power flow (and related problems) AI topics: Feasibility-constrained learning, graph neural networks, ML for optimization	Donti	DeepOPF: A Feasibility-Optimized Deep Neural Network Approach for AC Optimal Power Flow Problems CANOS: A Fast and Scalable Neural AC-OPF Solver Robust To N-1 Perturbations Learning to Solve Large-Scale Security-Constrained Unit Commitment Problems	DC3: A Learning Method for Optimization with Hard Constraints Other papers listed on ML OPF Wiki
Week 7 - Forked Lessons (Week 4 of Forked)
Mon 03/16	Climate Models	Synthesising & interpreting data sources; data assimilation	Bodner		ESM data downscaling: a comparison of super-resolution deep learning models
	Biodiversity	Long tails and anomalies	Beery	Multi-Scale and Multimodal Species Distribution Modeling BioCLIP: A Vision Foundation Model for the Tree of Life TaxaBind: A Unified Embedding Space for Ecological Applications
	Power & Energy	Paper presentations: Alternative approaches to optimal power flow (and related problems) AI topics: Neural ODEs, multi-agent learning, safe RL	Donti	Learning to Optimize meets Neural-ODE: Real-Time, Stability-Constrained AC OPF Towards Distributed Energy Services: Decentralizing Optimal Power Flow with Machine Learning RL2Grid: Benchmarking Reinforcement Learning in Power Grid Operations	Other papers listed on ML OPF Wiki
Wed 03/18	Climate Models	Synthesising & interpreting data sources; data assimilation	Bodner	Implementation and validation of a supermodeling framework into Community Earth System Model version 2.1.5 Downscaling daily climate data with super-resolution and CMIP6 global climate models to predict the climate in China High-resolution meteorology with climate change impacts from global climate model data using generative machine learning
	Biodiversity	Long tails and anomalies	Beery	BioCAP: Exploiting Synthetic Captions Beyond Labels in Biological Foundation Models ProM3E: Probabilistic Masked MultiModal Embedding Model for Ecology OlmoEarth: Stable Latent Image Modeling for Multimodal Earth Observation
	Power & Energy	Paper presentations: Generator control, voltage control, and demand response AI topics: Stability-constrained RL, multi-agent RL, multi-armed bandits	Donti	Stability Constrained Reinforcement Learning for Decentralized Real-Time Voltage Control Reinforcement learning for demand response: A review of algorithms and modeling techniques Bandit-based Solar Panel Control	The CityLearn Challenge 2022: Overview, Results, and Lessons Learned Helping mitigate climate change through efficient reinforcement learning-based wind farm flow control
Week 8 - Spring Break
Week 9 - Forked Lessons (Week 5 of Forked)
Mon 03/30	Climate Models	Foundation models and emulators of climate models	Bodner		Emulators of Climate Model Output
	Biodiversity	Multimodality	Beery	DISCount: Counting in Large Image Collections with Detector-Based Importance Sampling 21 000 birds in 4.5 h: efficient large-scale seabird detection with machine learning PPI: Prediction-Powered Inference
	Power & Energy	Lecture: Power Systems Planning	Donti
Wed 04/01	Climate Models	Foundation models and emulators of climate models	Bodner	SamudrACE: Fast and Accurate Coupled Climate Modeling with 3D Ocean and Atmosphere Emulators Aurora: A Foundation Model of the Atmosphere Neural general circulation models for modeling precipitation
	Biodiversity	Multimodality	Beery	Human-in-the-Loop Visual Re-ID for Population Size Estimation Conformal Prediction quantifies the uncertainty of Species Distribution Models Active Measurement: Efficient Estimation at Scale
	Power & Energy	Paper presentations: Power systems planning and scenario generation AI topics: Robust optimization, multi-objective optimization, generative modeling	Donti	A Deep Generative Learning Approach for Two-stage Adaptive Robust Optimization Efficiently Approximating the Pareto Frontier: Hydropower Dam Placement in the Amazon Basin Model-Free Renewable Scenario Generation Using Generative Adversarial Networks	EnhancedSD: Downscaling Solar Irradiance from Climate Model Projections
Week 10 - Forked Lessons (Week 6 of Forked)
Mon 04/06	Climate Models	Uncertainty quantification and estimating extreme events; scenario-dependence of future changes in extremes	Bodner		Artificial intelligence for modeling and understanding extreme weather and climate events Uncertainty Quantification of Satellite-Based Essential Climate Variables Derived from Deep Learning Parameter Uncertainty Quantification in an Idealized GCM With a Seasonal Cycle
	Biodiversity	Human-AI participation and integrating knowledge	Beery	Bridging adaptive management and reinforcement learning for more robust decisions Dataset Interfaces: Diagnosing Model Failures Using Controllable Counterfactual Generation Envisioning Communities: A Participatory Approach Towards AI for Social Good
	Power & Energy	Lecture: Accelerated Science, Knowledge Synthesis, and Data Synthesis	Donti
Wed 04/08	Climate Models	Uncertainty quantification and estimating extreme events; scenario-dependence of future changes in extremes	Bodner	Can AI weather models predict out-of-distribution gray swan tropical cyclones? Estimation of temperature and precipitation uncertainties using quantile neural networks Forecasting sea level maxima using Machine learning with explainability and extreme value analysis
	Biodiversity	Human-AI participation and integrating knowledge	Beery	Seeing Above and Below the Canopy: Modeling and Interpreting Species Occupancy with Multimodal Habitat Representations PAIR-X: Pairwise Matching of Intermediate Representations for Fine-grained Explainability Deep reinforcement learning for conservation decisions
	Power & Energy	Paper presentations: Accelerated science and knowledge synthesis AI topics: Physics-integrated ML, data-efficient learning, topic modeling	Donti	Accelerated Development of Perovskite-Inspired Materials via High-Throughput Synthesis and Machine-Learning Diagnosis Predicting disruptive instabilities in controlled fusion plasmas through deep learning Topic based classification and pattern identification in patents
Week 11 - Forked Lessons (Week 7 of Forked)
Mon 04/13 & Wed 04/15	All Tracks	Project Presentations	Bodner/Beery/Donti
Week 12 - Forked Lessons (Week 8 of Forked)
Mon (holiday) Wed 04/22	All Tracks	Project Presentations	Bodner/Beery/Donti
Week 13
Mon 04/27	Merged	Benchmarks & Evaluation	Beery
Wed 04/29	Merged	Data-centric research	Bodner
Week 14
Mon 05/04	Merged	Incorporation of domain knowledge	Donti
Wed 05/06	Merged	Guest Lecture: TBD	(Host: TBD)
Week 15
Mon 05/11	Merged	"Now what?" Opportunities to work on climate-related issues beyond the classroom: Moderated Panel	(Moderator: Beery)