The IMPACT team has two papers accepted at the ICLR 2024. The first Paper, WINDSET: Weather Insights and Novel Data for Systematic Evaluation and Testing is co-authored by IMPACT team members: Rajat Shinde, Christopher E Phillips, Sujit Roy, Manil Maskey, and Rahul Ramachandran (ST11). This paper introduces a set of benchmarks and evaluation framework for Artificial Intelligence (AI) applications for weather. The second paper, Clifford Neural Operators On Atmospheric Data Influenced Partial Differential Equations is co-authored by IMPACT team members: Roy, Wei Ji Leong, Shinde, Phillips, Ankur Kumar, Maskey, and Ramachandran. This study presents a novel approach of including geometric information to modeling partial differential equations that govern atmospheric dynamics using Clifford neural operators, offering significant improvement over traditional methods. Maskey is also organizing a specialized workshop focusing on data-centric machine learning research for science. The workshop is aimed at exploring the intersection of AI and scientific inquiry, particularly how data-centric AI approaches can be used to advance science. ICLR stands at the forefront of deep learning and representation learning research, offering a global platform for showcasing innovative research across a myriad of applications in artificial intelligence, statistics, data science, and more.
1 min read