Explore cutting-edge applications of deep learning in analyzing dissolved carbon dioxide levels in the Balearic coastal regions through this 22-minute lecture from the Instituto de Física Interdisciplinar y Sistemas Complejos (IFISC). Delve into advanced concepts such as self-attention, permutation convolution, and symbolic regression, while examining the results through principal component analysis. Gain valuable insights into the intersection of machine learning and environmental science, and understand how these techniques contribute to our understanding of coastal ecosystems and climate change impacts.