How Machine Learning can join the battle against climate change
On a sunny Wednesday evening, 70 people from all across the globe gathered together in our online think tank to put their minds on global warming and machine learning (ML).
On the menu: 3 renowned speakers, Q&A sessions and breakout discussion rooms. Our own AI-lead started off by pointing the noses in the same direction. He mentioned that AI & ML are just one tool in a broader toolkit that can be used to support or accelerate climate change strategies.
It’s no silver bullet solution, but combining knowledge & expertise can have a significant impact on trying to keep our planet healthy.
Machine Learning against global warming
Evan D. Sherwin was the first to address the crowd. As a data-informed energy policy analyst and co-founder of CCAI (Climate Change AI), he brought well-earned expertise and insights to the stage. Sherwin introduced us to the countless opportunities of Machine Learning to help tackle climate change. He and his collaborators joined efforts to create a 100-page report on ways how ML might be useful in the battle against global warming.
Without diving into specifics, Sherwin managed to highlight some key takeaways of his report. He mainly mentioned the strong points of ML and linked them to possible use cases:
- Distilling observed data: leverage ML to analyze satellite images in order to spot the emission of greenhouse gases.
- Improving efficiency: ML can help optimize demand response on electrical grids, making the grid as efficient as possible.
- Forecasting: using time-series predictions to battle overproduction and waste.
- Accelerating simulations: train algorithms to replace the time-consuming physical models.
- Accelerated experimentation: ML can give suggestions for which experiments to try, based on experiments we’ve done. It can show us the right direction in experimenting with new green materials.
Perhaps most important of all, Sherwin used his contagious enthusiasm to call everyone out to work together, take initiatives and share knowledge across domains. It won’t come as a surprise if the CCAI will have a few more contributors in the next few days.
Putting it into practice
In the spirit of building towards a sustainable future, Thomas Vrancken from ML6 and Barack Chizi from KBC presented some strong use cases on the edge of climate and machine learning.
ML6 managed to increase the efficiency of the DeepWind offshore wind turbine park by thorough data exploration and root cause analysis. They proved that data science has strong potential for making new energy sources more productive.
KBC, on the other hand, talked about their corporate social responsibility and showcased some of their many initiatives to live up to the Paris agreements by investigating the ecological footprint of investments, credits and loans through AI.
What sticks with us after an evening of ML-brainiacs sharing the stage? It’s not that ML will somehow eradicate global warming. Again, it’s not a silver bullet solution. But if we bring the core strengths of Machine Learning into the picture, share our learnings and build bridges between engineers, data analysts and climate scientists, machine learning can be a most valuable tool in the united fight against climate change.