Thank you for joining the 1st AI for Earth Sciences workshop virtually at ICLR and making it a success in these challenging times!!


AI for Earth Sciences will also be joining Tackling Climate Change with ML on the 29th of April for a Climate Science and Adaptation Day between 7am-12.30pm PDT for sessions on Climate Change Adaptability, Emulators and Forecasting, click here for zoom registration details. You may use our slack channel ai4earth, #askauthor for Q/A similar to our workshop on April 26th.

For those who missed our workshop, you can catch the full recorded webinar, papers & presentations below


Our workshop brings together Earth scientists and machine learning experts to try to solve some of the Earth’s greatest problems. We’ve divided our workshop into several themed sections: Atmospheric Science, Hydro and Cryospheres, Solid Earth, Theoretical Advances, Remote Sensing, EnviroNet, Keynotes.


Tune into our livestream on April 26 from 7am-4pm PDT (San Francisco) time to see the talks. Join us in slack to mingle with workshop attendees.

The duration shown in the schedule is approximate time allocated for introduction, video recording, and Q&A for each topic.

Zoom attendees must completely close the ICLR stream before speaking in the zoom conference (seriously - you will regret it if you don’t)

All times are listed in Pacific Daylight Time (San Francisco, USA).

Start End Type Speaker Title
7:00 7:03 Welcome Organising Team AI for Earth Sciences

Atmospheric Science

Start End Type Speaker & Video Title
7:03 7:20 Invited Amy McGovern Using Machine Learning And Model Interpretation And Visualization Techniques To Gain Physical Insights In Atmospheric Science
7:20 7:40 Spotlight Haolin Fei Accurate Air Quality Prediction: A Physical-temporal Collection Model
7:40 7:50 Lightning Jing Li A Random Forest Model For The Probability Of Large Wildfires In California
7:50 8:00 Lightning Ashray Manepalli Generalization Properties Of Machine Learning Based Weather Model Downscaling
8:00 8:10 Lightning Adway Mitra A Probabilistic Graphical Model Approach To Identifying Spatial Changes In Monthly Precipitation Under Climate Change

Hydro and Cryospheres

Start End Type Speaker & Video Link Title and Paper Link
8:10 8:25 Invited Kelly Kochanski Surrogate Sea Ice Model Enables Efficient Tuning
8:25 8:48 Invited Zach Moshe Hydronets: Leveraging River Structure for Hydrologic Modeling
8:48 9:00 Lightning Brian Cerrón Detection Of Housing And Agriculture Areas On Dry-riverbeds For The Evaluation Of Risk By Landslides Using Low-resolution Satellite Imagery Based On Deep Learning. Study Zone: Lima, Peru
XX XX Abstract Mearg Belay B. Shibeshi Geo-spatial Approach For Assessing The Impact Of Land-use And Land-cover Change On Groundwater Recharge: A Case Study In Akaki Catchment, Central Ethiopia
9:00 9:10 Morning Break    

Solid Earth

Start End Type Speaker & Video Link Title & Paper Link
9:10 9:30 Spotlight Seyed M Mousavi Hierarchical Attentive Modeling Of Earthquake Signals
9:30 9:48 Regular Bas Peters Fully Reversible Neural Networks For Large-scale Surface And Sub-surface Characterization Via Remote Sensing
9:48 10:00 Lightning Tue Boesen Semi-supervised Clustering For Oil Prospectivity
XX XX Abstract Hadeer A El Ashhab Modeling Hydrocarbons Flow From Earth Using Deep Learning

Theoretical Advances

Start End Type Speaker & Video Link Title & Paper Link
10:00 10:30 Spotlight Arvind T Mohan Embedding Hard Physical Constraints In Convolutional Neural Networks For 3d Turbulence
10:30 10:55 Spotlight Arvind T Mohan Wavelet-powered Neural Networks For Turbulence
10:55 11:15 Regular Srija Chakraborty Time-varying Semantic Representations Of Planetary Observations For Discovering Novelties
XX XX Abstract Taesung Kim Gaganet: End-to-end Multivariate Time Series Imputation And Prediction With Gated Generated Adversarial Networks
11:15 11:30 Noon Break    

Remote Sensing

Start End Type Speaker & Video Link Title & Paper Link
11:30 12pm Invited Ethan Weber & Hassan Kane Building Disaster Damage Assessment in Satellite Imagery with Multi-Temporal Fusion
12pm 12:30 Spotlight Maxim Neumann In-domain Representation Learning For Remote Sensing


Start End Type Speaker & Video Link Title & Paper Link
12:30 13:00 Invited Lukas Kapp-Schwoerer Climatenet: Bringing The Power Of Deep Learning To Weather And Climate Sciences Via Open Datasets And Architectures
13:00 13:15 Invited Stephan Rasp WeatherBench: A benchmark dataset for data-driven weather forecasting
13:15 13:30 Lightning Seyed M Mousavi Dataset Labeling Paper: AI-based Earthquake Signal Detection And Processing
13:30 13:55 Invited Edward W Obropta Infrared Solar Module Dataset For Anomaly Detection
13:55 14:15 Invited Ankush Khandelwal AquaNet
14:15 14:25 Extended Discussion Vipin Kumar & S. Karthik Mukkavilli EnviroNet

Keynotes & Discussion

Start End Type Speaker & Video Link Title & Website
14:30 15:00 Keynote Prof. Daniel M. Kammen, UC Berkeley Data Science for the Clean Energy Revolution
15:00 15:30 Discussion Keynote & AI for Earth Sciences Team AI Synergies in Energy, Resources & Earth System
15:30 16:00 Closing Keynote Paul Miller aka DJ Spooky Art, AI & Earth Sciences

Contact Us

Send inquiries to ai4earthscience[at]gmail[dot]com