|Researchers are leveraging machine learning to mitigate the impacts of climate change / Credits: muratart via Shutterstock|
In last year’s Conference and Workshop on Neural Information Processing Systems (NeurIPS), a machine learning and computational neuroscience conference held every December, experts discussed how artificial intelligence and machine learning can help in addressing one of humanity’s greatest threats: climate change. This is part of an effort to tackle how modern technologies can save people from climate change’s impact.
The “Tackling Climate Change” workshop focuses on two things: encouraging machine learning practitioners to use their research to solve climate change problems and ensuring that those studies will not ironically contribute to climate change. ML leaders such as Yoshua Bengio, Jeff Dean, Andrew Ng, and Carla Gomes have agreed that ML researchers should focus on solving huge challenges like climate change instead of worrying about the number of publications they are getting.
According to VentureBeat, the leader in covering transformative tech that helps business leaders make smarter decisions with our industry-leading AI and gaming coverage, the ML leaders suggested that ML should be used more energy efficiency. This includes using transfer learning, self-supervised learning, and multitask learning and achieving accurate results using smaller data sets. At the same time, the experts also noted how studies have burned a great deal of energy since running models is computationally intensive.
“Some of these large models are computationally intensive, and they’re reasonably expensive in terms of energy usage. And so I think it’s important for the community to look at what are more efficient algorithmic techniques we can have that make a particular model or outcome that we want,” Dean said.
Also, the study “Quantifying the Carbon Emissions of Machine Learning” developed a tool that will help researchers determine how much carbon will their project emit and how much (if any) of that is offset by the provider. The researchers created ML CO2 Impact, a machine learning emissions calculator, which can be a useful tool for those working in ML and also a philosophical reminder of the responsibilities they bear for their work.