🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Our new preprint on a probabilistic multi-fidelity #climate model parameterization to extrapolate to unseen warmer climates is out! Moving towards better climate change projections with #ArtificialIntelligence Liran Peng Mike Pritchard Pierre Gentine arxiv.org/abs/2309.10231

Our new CMAME paper on scalable Bayesian optimization with randomized prior networks is out! The method accounts for constrained multi-fidelity optimization tasks & is applied to compressor's blades shape optimization in turbo-machinery. Paris Perdikaris sciencedirect.com/science/articl…

I’ll attend NeurIPS next week in New Orleans, where I’ll present ClimSim, the huge dataset developed to address a key challenge in hybrid-ML climate modeling. Don’t hesitate to reach out for a chat if you'll be at NeurIPS and are curious about ML challenges in climate science!!🌎

**Outstanding Datasets and Benchmarks Papers** ClimSim: A large multi-scale dataset for hybrid physics-ML climate emulation DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models

THIS WEEK is the 1st Spring'24 Lecture in Climate Data Science: Research Updates from Kaitlyn Loftus + Mohamed Aziz Bhouri! 📅 THURS 1/18 @ 12p (EDT) 🗺️ In-person / Zoom 💻 RSVP: bit.ly/3vEUATK #climate #model #ML #cloud #physics #data Columbia Climate School Columbia University Columbia Engineering U.S. National Science Foundation

i haven't used a debugger in 2 years

I am happy to announce that I will be in Vienna from 17 to 19 april, to participate in #EGU24! I will make an oral presentation about the coupling between ARPEGE-climat and a NN parameterization. This is part of my work from last fall in Columbia. meetingorganizer.copernicus.org/EGU24/session/…

In this article, accepted at ICLR2024, Yongquan Qu, Mohamed Aziz Bhouri , and Pierre Gentine, tackle the joint inference and uncertainty quantification of poorly known physical parameters and machine learning parametrizations, for sub-grid scale dynamics. doi.org/10.48550/arXiv…

This is our first step towards modernizing parameterization/data assimilation. We believe that differentiable programming can unlock new potentials in #AI4Science research. Looking forward to discussions at ICLR2024 workshop on AI4DifferentialEquations in Science!