S. Karthik Mukkavilli

Climate change is one of the greatest challenges facing humanity, and we, as machine learning experts, may wonder how we can help. Here we describe how machine learning can be a powerful tool in reducing greenhouse gas emissions and helping society adapt to a changing climate. From smart grids to disaster management, we identify high impact problems where existing gaps can be filled by machine learning, in collaboration with other fields. Our recommendations encompass exciting research questions

We present a project that aims to generate images that depict accurate,\nvivid, and personalized outcomes of climate change using Cycle-Consistent\nAdversarial Networks (CycleGANs). By training our CycleGAN model on street-view\nimages of houses before and after extreme weather events (e.g. floods, forest\nfires, etc.), we learn a mapping that can then be applied to images of\nlocations that have not yet experienced these events. This visual\ntransformation is paired with climate model predictio

Reinforcement learning is emerging as a powerful tool for microrobots control, as it enables autonomous navigation in environments where classical control approaches fall short. However, applying reinforcement learning to microrobotics is difficult due to the need for large training datasets, the slow convergence in physical systems and poor generalizability across environments. These challenges are amplified in ultrasound-actuated microrobots, which require rapid, precise adjustments in high-di

Abstract Floods, droughts, and rainfall‐induced landslides are hydro‐hazards that affect millions of people every year. Anticipation, mitigation, and adaptation to these hazards is increasingly outpaced by their changing magnitude and frequency due to climate change. A key question for society is whether the research we pursue has the potential to address knowledge gaps and to reduce potential future hazard impacts where they will be most severe. We use natural language processing, based on a ne

Abstract Direct normal irradiance (DNI) is the main input for concentrating solar power (CSP) technologies—an important component in future energy scenarios. DNI forecast accuracy is sensitive to radiative transfer schemes (RTSs) and microphysics in numerical weather prediction (NWP) models. Additionally, NWP models have large regional aerosol uncertainties. Dust aerosols can significantly attenuate DNI in extreme cases, with marked consequences for applications such as CSP. To date, studies hav

Machine learning and deep learning methods have been widely explored in understanding the chaotic behavior of the atmosphere and furthering weather forecasting. There has been increasing interest from technology companies, government institutions, and meteorological agencies in building digital twins of the Earth. Recent approaches using transformers, physics-informed machine learning, and graph neural networks have demonstrated state-of-the-art performance on relatively narrow spatiotemporal sc

Abstract Increasing publication numbers make it difficult to keep up with knowledge evolution in a science like hydrology. Here we give recommendations to authors and journals for writing future‐proof articles that contribute to knowledge accumulation and synthesis.

Significant progress in the development of highly adaptable and reusable Artificial Intelligence (AI) models is expected to have a significant impact on Earth science and remote sensing. Foundation models are pre-trained on large unlabeled datasets through self-supervision, and then fine-tuned for various downstream tasks with small labeled datasets. This paper introduces a first-of-a-kind framework for the efficient pre-training and fine-tuning of foundational models on extensive geospatial dat

Bishwaranjan Bhattacharjee, Aashka Trivedi, Masayasu Muraoka, Muthukumaran Ramasubramanian, Takuma Udagawa, Iksha Gurung, Nishan Pantha, Rong Zhang, Bharath Dandala, Rahul Ramachandran, Manil Maskey, Kaylin Bugbee, Michael M. Little, Elizabeth Fancher, Irina Gerasimov, Armin Mehrabian, Lauren Sanders, Sylvain V. Costes, Sergi Blanco-Cuaresma, Kelly Lockhart, Thomas Allen, Felix Grezes, Megan Ansdell, Alberto Accomazzi, Yousef El-Kurdi, Davis Wertheimer, Birgit Pfitzmann, Cesar Berrospi Ramis, Mi

Though machine learning has achieved notable success in modeling sequential and spatial data for speech recognition and in computer vision, applications to remote sensing and climate science problems are seldom considered. In this paper, we demonstrate techniques from unsupervised learning of future video frame prediction, to increase the accuracy of ice flow tracking in multi-spectral satellite images. As the volume of cryosphere data increases in coming years, this is an interesting and import

Floods, droughts, and rainfall-induced landslides are hydro-geomorphic hazards that affect millions of people every year. Anticipation, mitigation, and adaptation to these hazards is increasingly outpaced by their changing magnitude and frequency due to climate change. A key question for society is whether the research we pursue has the potential to address knowledge gaps and to reduce potential future hazard impacts where they will be the most severe. We use natural language processing, based o

Abstract AI has catalyzed transformative advancements across multiple sectors, from medical diagnostics to autonomous vehicles, enhancing precision and efficiency. As it ventures into microrobotics, AI offer innovative solutions to the formidable challenge of controlling and manipulating microrobots, which typically operate within imprecise, remotely actuated systems—a task often too complex for human operators. We implement state-of-the-art model-based reinforcement learning for autonomous cont