LEARNING OUTCOMES
The purpose of this special course is to demonstrate how machine learning methods can be used to tackle climate change.
Credits: 5
Schedule: 08.01.2024 - 15.02.2024
Teacher in charge (valid for whole curriculum period):
Teacher in charge (applies in this implementation): Olga Kuznetsova, Laia Amoros Carafi
Contact information for the course (applies in this implementation):
Laia Amorós (Finnish Meteorological Institute), Olga Kuznetsova (Aalto University), Shaikhum Monira (Aalto University, shaikhum.monira@aalto.fi)
CEFR level (valid for whole curriculum period):
Language of instruction and studies (applies in this implementation):
Teaching language: English. Languages of study attainment: English
CONTENT, ASSESSMENT AND WORKLOAD
Workload
applies in this implementation
24 hours contact teaching (12 hours lectures and 12 hours of exercise sessions) and 111 hours of independent work. Overall, the idea is that after introducing the course and data in the first week, we focus on one policy area per week (e.g., energy efficiency measure in buildings and cities), discuss its importance and potential for climate action and then critically analyse the methodology and findings of a research article where ML was applied to that policy area. In the exercise session of the same week, we create an implementation of an ML method from the same family of algorithms (e.g., computer vision) to real-life data coming from the same policy area.
DETAILS
Study Material
applies in this implementation
- ML textbook
- ML for climate action
- Research articles for each lecture:
- Lecture II: D. P. Finch, P. I. Palmer, and T. Zhang (2022). Automated detection of atmospheric NO2 plumes from satellite data: a tool to help infer anthropogenic combustion emissions. Atmospheric Measurement Techniques, 15(3):721–733, 2022.
- Lecture III: Schuit, B. J., Maasakkers, J. D., Bijl, P., Mahapatra, G., Van den Berg, A. W., Pandey, S., Aben, I. (2023). Automated detection and monitoring of methane super-emitters using satellite data. Atmospheric Chemistry and Physics Discussions, 1-47.
- Lecture IV: Paolo, F., Kroodsma, D., Raynor, J. et al. Satellite mapping reveals extensive industrial activity at sea. Nature 625, 85–91 (2024). https://doi.org/10.1038/s41586-023-06825-8
- Lecture V:
- Lecture VI:
Substitutes for Courses
valid for whole curriculum period:
Prerequisites
valid for whole curriculum period:
FURTHER INFORMATION
Further Information
valid for whole curriculum period:
This is aimed at students who have some previous knowledge of Python and the main machine learning algorithms (at the level of the course CS-EJ3211 "Machine Learning with Python"). We will study relevant literature on different problems such as computer vision for remote sensing of farms, forests and emissions; causal inference for improving energy efficiency of buildings and cities, time-series forecasting for managing transport emissions, and unsupervised learning in electricity systems. Students will apply the methods to real data from satellites and other data sources.
Target audience: primary - Master's in Machine Learning, Data Science and Artificial Intelligence; secondary: Master's in Business Analytics and Mathematics and Operations Research.
Details on the schedule
applies in this implementation
CS-E407519 Special Course in Machine Learning, Data Science and Artificial Intelligence: Machine Learning for Climate Action
Class Schedule:
- Mon 08.01.2024 12:15 - 14:00, R001/Y313
- Mon 15.01.2024 12:15 - 14:00, R001/Y313
- Mon 22.01.2024 12:15 - 14:00, R001/Y313
- Mon 29.01.2024 12:15 - 14:00, R001/Y313
- Mon 05.02.2024 12:15 - 14:00, R001/Y313
- Mon 12.02.2024 12:15 - 14:00, R001/Y313
Exercise Session:
- Thu 11.01.2024 10:15 - 12:00, R037/1521-1522 AS6
- Thu 18.01.2024 10:15 - 12:00, R030/A136 T6
- Thu 25.01.2024 10:15 - 12:00, R037/1521-1522 AS6
- Thu 01.02.2024 10:15 - 12:00, R030/A136 T6
- Thu 08.02.2024 10:15 - 12:00, R030/A136 T6
- Thu 15.02.2024 10:15 - 12:00, R030/A136 T6