Opening the neural networks’ black box for climate science | Pedram Hassanzadeh, Rice University
0The global climate adaptation and mitigation efforts require reliable information about the future of climate variability and extremes, particularly at regional scales. Integrating observations, theory, and physics-based computer models has resulted in significant advances in climate science in the past century. However, there are still major challenges in obtaining reliable, actionable information due to the large uncertainties in climate projections. Thus, there is a critical need for drastic improvements in our fundamental understanding of the Earth system and modeling capabilities. Here, I will argue that integrating scientific machine learning (ML) with the conventional approach could potentially open new avenues to substantially accelerate climate research, e.g., via developing better and faster weather/climate models, extracting more information from observational data, and even (potentially) improving our fundamental knowledge. However, as scientific ML is in its infancy, there are major challenges for climate applications that need to be addressed first. These challenges include interpretability, stability, extrapolation, and learning in the small-data regime. The talk will highlight some of the recent work on the promises, challenges, and future possibilities of applying scientific ML to accelerate climate research. In particular, a new method will be discussed, based on integrating the Fourier analyses of neural networks and climate data, which enables to explain and connect the learned physics and inner workings of the network. This method is a step toward developing a much-needed general framework to rigorously analyze and understand neural networks for climate applications and making them reliable and effective tools. \n\nSpeakers:\n\nPedram Hassanzadeh\nRice University\n\nModerators:\nDuncan Watson-Parris\nUniversity of California San Diego\n\nPhilip Stier\nUniversity of Oxford\n\nJoin us for two days of never before presented, state of the art AI solutions and cutting edge knowledge, aligned with the UN Sustainable Development Goals.\nRegister and learn more here: https://aiforgood.itu.int/summit23/\n\nJoin the Neural Network!\nhttps://aiforgood.itu.int/neural-network/\nThe AI for Good networking community platform powered by AI. \nDesigned to help users build connections with innovators and experts, link innovative ideas with social impact opportunities, and bring the community together to advance the SDGs using AI.\n\n Watch the latest #AIforGood videos!\n\n\n Stay updated and join our weekly AI for Good newsletter:\nhttp://eepurl.com/gI2kJ5\n\nCheck out the latest AI for Good news:\nhttps://aiforgood.itu.int/newsroom/\n\nExplore the AI for Good blog:\nhttps://aiforgood.itu.int/ai-for-good-blog/\n\n Connect on our social media:\nWebsite: https://aiforgood.itu.int/\nTwitter: https://twitter.com/AIforGood\nLinkedIn Page: https://www.linkedin.com/company/26511907 \nLinkedIn Group: https://www.linkedin.com/groups/8567748 \nInstagram: https://www.instagram.com/aiforgood \nFacebook: https://www.facebook.com/AIforGood\n\nWhat is AI for Good?\nWe have less than 10 years to solve the UN SDGs and AI holds great promise to advance many of the sustainable development goals and targets.\nMore than a Summit, more than a movement, AI for Good is presented as a year round digital platform where AI innovators and problem owners learn, build and connect to help identify practical AI solutions to advance the United Nations Sustainable Development Goals.\nAI for Good is organized by ITU in partnership with 40 UN Sister Agencies and co-convened with Switzerland.\n\nDisclaimer:\nThe views and opinions expressed are those of the panelists and do not reflect the official policy of the ITU.