Björn Lütjens, Catherine H. Crawford, Mark Veillette, Dava Newman
Climate models project an uncertainty range of possible warming scenarios from 1.5 to 5 degree Celsius global temperature increase until 2100, according to the CMIP6 model ensemble. Climate risk management and infrastructure adaptation requires the accurate quantification of the uncertainties at the local level. Ensembles of high-resolution climate models could accurately quantify the uncertainties, but most physics-based climate models are computationally too expensive to run as ensemble. Recent works in physics-informed neural networks (PINNs) have combined deep learning and the physical sciences to learn up to 15k faster copies of climate submodels. However, the application of PINNs in climate modeling has so far been mostly limited to deterministic models. We leverage a novel method that combines polynomial chaos expansion (PCE), a classic technique for uncertainty propagation, with PINNs. The PCE-PINNs learn a fast surrogate model that is demonstrated for uncertainty propagation of known parameter uncertainties. We showcase the effectiveness in ocean modeling by using the local advection-diffusion equation.
Ziv Epstein, Hope Schroeder, Dava Newman
Generative AI techniques like those that synthesize images from text (text-to-image models) offer new possibilities for creatively imagining new ideas. We investigate the capabilities of these models to help communities engage in conversations about their collective future. In particular, we design and deploy a facilitated experience where participants collaboratively speculate on utopias they want to see, and then produce AI-generated imagery from those speculations. In a series of in-depth user interviews, we invite participants to reflect on the generated images and refine their visions for the future. We synthesize findings with a bespoke community zine on the experience. We observe that participants often generated ideas for implementing their vision and drew new lateral considerations as a result of viewing the generated images. Critically, we find that the unexpected difference between the participant's imagined output and the generated image is what facilitated new insight for the participant. We hope our experimental model for co-creation, computational creativity, and community reflection inspires the use of generative models to help communities and organizations envision better futures.
Björn Lütjens, Brandon Leshchinskiy, Christian Requena-Mesa, Farrukh Chishtie, Natalia Díaz-Rodriguez, Océane Boulais, Aaron Piña, Dava Newman, Alexander Lavin, Yarin Gal, Chedy Raïssi
As climate change increases the intensity of natural disasters, society needs better tools for adaptation. Floods, for example, are the most frequent natural disaster, but during hurricanes the area is largely covered by clouds and emergency managers must rely on nonintuitive flood visualizations for mission planning. To assist these emergency managers, we have created a deep learning pipeline that generates visual satellite images of current and future coastal flooding. We advanced a state-of-the-art GAN called pix2pixHD, such that it produces imagery that is physically-consistent with the output of an expert-validated storm surge model (NOAA SLOSH). By evaluating the imagery relative to physics-based flood maps, we find that our proposed framework outperforms baseline models in both physical-consistency and photorealism. While this work focused on the visualization of coastal floods, we envision the creation of a global visualization of how climate change will shape our earth.
Peishi Jiang, Nis Meinert, Helga Jordão, Constantin Weisser, Simon Holgate, Alexander Lavin, Björn Lütjens, Dava Newman, Haruko Wainwright, Catherine Walker, Patrick Barnard
Developing fast and accurate surrogates for physics-based coastal and ocean models is an urgent need due to the coastal flood risk under accelerating sea level rise, and the computational expense of deterministic numerical models. For this purpose, we develop the first digital twin of Earth coastlines with new physics-informed machine learning techniques extending the state-of-art Neural Operator. As a proof-of-concept study, we built Fourier Neural Operator (FNO) surrogates on the simulations of an industry-standard flood and ocean model (NEMO). The resulting FNO surrogate accurately predicts the sea surface height in most regions while achieving upwards of 45x acceleration of NEMO. We delivered an open-source \textit{CoastalTwin} platform in an end-to-end and modular way, to enable easy extensions to other simulations and ML-based surrogate methods. Our results and deliverable provide a promising approach to massively accelerate coastal dynamics simulators, which can enable scientists to efficiently execute many simulations for decision-making, uncertainty quantification, and other research activities.
Björn Lütjens, Catherine H. Crawford, Campbell D Watson, Christopher Hill, Dava Newman
Numerical simulations in climate, chemistry, or astrophysics are computationally too expensive for uncertainty quantification or parameter-exploration at high-resolution. Reduced-order or surrogate models are multiple orders of magnitude faster, but traditional surrogates are inflexible or inaccurate and pure machine learning (ML)-based surrogates too data-hungry. We propose a hybrid, flexible surrogate model that exploits known physics for simulating large-scale dynamics and limits learning to the hard-to-model term, which is called parametrization or closure and captures the effect of fine- onto large-scale dynamics. Leveraging neural operators, we are the first to learn grid-independent, non-local, and flexible parametrizations. Our \textit{multiscale neural operator} is motivated by a rich literature in multiscale modeling, has quasilinear runtime complexity, is more accurate or flexible than state-of-the-art parametrizations and demonstrated on the chaotic equation multiscale Lorenz96.
Alexandre Lacoste, Nils Lehmann, Pau Rodriguez, Evan David Sherwin, Hannah Kerner, Björn Lütjens, Jeremy Andrew Irvin, David Dao, Hamed Alemohammad, Alexandre Drouin, Mehmet Gunturkun, Gabriel Huang, David Vazquez, Dava Newman, Yoshua Bengio, Stefano Ermon, Xiao Xiang Zhu
Recent progress in self-supervision has shown that pre-training large neural networks on vast amounts of unsupervised data can lead to substantial increases in generalization to downstream tasks. Such models, recently coined foundation models, have been transformational to the field of natural language processing. Variants have also been proposed for image data, but their applicability to remote sensing tasks is limited. To stimulate the development of foundation models for Earth monitoring, we propose a benchmark comprised of six classification and six segmentation tasks, which were carefully curated and adapted to be both relevant to the field and well-suited for model evaluation. We accompany this benchmark with a robust methodology for evaluating models and reporting aggregated results to enable a reliable assessment of progress. Finally, we report results for 20 baselines to gain information about the performance of existing models. We believe that this benchmark will be a driver of progress across a variety of Earth monitoring tasks.
Alexander Lavin, Ciarán M. Gilligan-Lee, Alessya Visnjic, Siddha Ganju, Dava Newman, Atılım Güneş Baydin, Sujoy Ganguly, Danny Lange, Amit Sharma, Stephan Zheng, Eric P. Xing, Adam Gibson, James Parr, Chris Mattmann, Yarin Gal
The development and deployment of machine learning (ML) systems can be executed easily with modern tools, but the process is typically rushed and means-to-an-end. The lack of diligence can lead to technical debt, scope creep and misaligned objectives, model misuse and failures, and expensive consequences. Engineering systems, on the other hand, follow well-defined processes and testing standards to streamline development for high-quality, reliable results. The extreme is spacecraft systems, where mission critical measures and robustness are ingrained in the development process. Drawing on experience in both spacecraft engineering and ML (from research through product across domain areas), we have developed a proven systems engineering approach for machine learning development and deployment. Our "Machine Learning Technology Readiness Levels" (MLTRL) framework defines a principled process to ensure robust, reliable, and responsible systems while being streamlined for ML workflows, including key distinctions from traditional software engineering. Even more, MLTRL defines a lingua franca for people across teams and organizations to work collaboratively on artificial intelligence and machine learning technologies. Here we describe the framework and elucidate it with several real world use-cases of developing ML methods from basic research through productization and deployment, in areas such as medical diagnostics, consumer computer vision, satellite imagery, and particle physics.
Rupa Kurinchi-Vendhan, Björn Lütjens, Ritwik Gupta, Lucien Werner, Dava Newman
The transition to green energy grids depends on detailed wind and solar forecasts to optimize the siting and scheduling of renewable energy generation. Operational forecasts from numerical weather prediction models, however, only have a spatial resolution of 10 to 20-km, which leads to sub-optimal usage and development of renewable energy farms. Weather scientists have been developing super-resolution methods to increase the resolution, but often rely on simple interpolation techniques or computationally expensive differential equation-based models. Recently, machine learning-based models, specifically the physics-informed resolution-enhancing generative adversarial network (PhIREGAN), have outperformed traditional downscaling methods. We provide a thorough and extensible benchmark of leading deep learning-based super-resolution techniques, including the enhanced super-resolution generative adversarial network (ESRGAN) and an enhanced deep super-resolution (EDSR) network, on wind and solar data. We accompany the benchmark with a novel public, processed, and machine learning-ready dataset for benchmarking super-resolution methods on wind and solar data.
Björn Lütjens, Brandon Leshchinskiy, Océane Boulais, Farrukh Chishtie, Natalia Díaz-Rodríguez, Margaux Masson-Forsythe, Ana Mata-Payerro, Christian Requena-Mesa, Aruna Sankaranarayanan, Aaron Piña, Yarin Gal, Chedy Raïssi, Alexander Lavin, Dava Newman
Deep generative vision models are now able to synthesize realistic-looking satellite imagery. But, the possibility of hallucinations prevents their adoption for risk-sensitive applications, such as generating materials for communicating climate change. To demonstrate this issue, we train a generative adversarial network (pix2pixHD) to create synthetic satellite imagery of future flooding and reforestation events. We find that a pure deep learning-based model can generate photorealistic flood visualizations but hallucinates floods at locations that were not susceptible to flooding. To address this issue, we propose to condition and evaluate generative vision models on segmentation maps of physics-based flood models. We show that our physics-conditioned model outperforms the pure deep learning-based model and a handcrafted baseline. We evaluate the generalization capability of our method to different remote sensing data and different climate-related events (reforestation). We publish our code and dataset which includes the data for a third case study of melting Arctic sea ice and $>$30,000 labeled HD image triplets -- or the equivalent of 5.5 million images at 128x128 pixels -- for segmentation guided image-to-image translation in Earth observation. Code and data is available at \url{https://github.com/blutjens/eie-earth-public}.
Daniel Platnick, Mohamed E. Bengueddache, Marjan Alirezaie, Dava J. Newman, Alex ''Sandy'' Pentland, Hossein Rahnama
Generative agents powered by language models are increasingly deployed for long-horizon tasks. However, as long-term memory context grows over time, they struggle to maintain coherence. This deficiency leads to critical failures, including identity drift, ignoring established beliefs, and the propagation of hallucinations in multi-agent systems. To mitigate these challenges, this paper introduces Identity Retrieval-Augmented Generation (ID-RAG), a novel mechanism designed to ground an agent's persona and persistent preferences in a dynamic, structured identity model: a knowledge graph of core beliefs, traits, and values. During the agent's decision loop, this model is queried to retrieve relevant identity context, which directly informs action selection. We demonstrate this approach by introducing and implementing a new class of ID-RAG enabled agents called Human-AI Agents (HAis), where the identity model is inspired by the Chronicle structure used in Perspective-Aware AI, a dynamic knowledge graph learned from a real-world entity's digital footprint. In social simulations of a mayoral election, HAis using ID-RAG outperformed baseline agents in long-horizon persona coherence - achieving higher identity recall across all tested models by the fourth timestep - and reduced simulation convergence time by 19% (GPT-4o) and 58% (GPT-4o mini). By treating identity as an explicit, retrievable knowledge structure, ID-RAG offers a foundational approach for developing more temporally coherent, interpretable, and aligned generative agents. Our code is open-source and available at: https://github.com/flybits/humanai-agents.
Daniel Platnick, Matti Gruener, Marjan Alirezaie, Kent Larson, Dava J. Newman, Hossein Rahnama
AI-enhanced Extended Reality (XR) aims to deliver adaptive, immersive experiences-yet current systems fall short due to shallow user modeling and limited cognitive context. We introduce Perspective-Aware AI in Extended Reality (PAiR), a foundational framework for integrating Perspective-Aware AI (PAi) with XR to enable interpretable, context-aware experiences grounded in user identity. PAi is built on Chronicles: reasoning-ready identity models learned from multimodal digital footprints that capture users' cognitive and experiential evolution. PAiR employs these models in a closed-loop system linking dynamic user states with immersive environments. We present PAiR's architecture, detailing its modules and system flow, and demonstrate its utility through two proof-of-concept scenarios implemented in the Unity-based OpenDome engine. PAiR opens a new direction for human-AI interaction by embedding perspective-based identity models into immersive systems.