| https://www.linkedin.com/in/benblaiszik |
🛫 A new generation of scientific understanding and #AI is on the horizon with the Exascale Aurora system. Very nice article profiling the efforts of Rick Stevens, Michael Papka, and the Argonne Leadership Computing Facility team.
“You’ve got a community of people that have been working on advancing high-performance computing for decades. And it powers the whole economy.” - Rick Stevens
Article: https://lnkd.in/g2s3w-U5
Registration is now open for the Materials Research Data Alliance (MaRDA) annual conference! The conference will be held fully virtual with free registration (Feb 21-23). List of speakers forthcoming soon, but I can guarantee there are exciting people involved and the discussions will be great.
Registration: https://lnkd.in/g5gyZuUy
#ai #ml #openscience #data #materialsscience #acceleratedscience
New study from Globus Labs uses #ML to classify clouds in satellite images to better understand environmental dynamics. 🌪️🌤️🌩️ #science #environment
Link: https://doi.org/10.3390/rs14225690
Globus Labs Research Group: https://labs.globus.org
Authors: Takuya Kurihana, Ian Foster, Liz Moyer
University of Chicago, Argonne National Laboratory, Argonne Leadership Computing Facility
Clouds play an important role in the Earth’s energy budget, and their behavior is one of the largest uncertainties in future climate projections. Satellite observations should help in understanding cloud responses, but decades and petabytes of multispectral cloud imagery have to date received only limited use. This study describes a new analysis approach that reduces the dimensionality of satellite cloud observations by grouping them via a novel automated, unsupervised cloud classification technique based on a convolutional autoencoder, an artificial intelligence (AI) method good at identifying patterns in spatial data. Our technique combines a rotation-invariant autoencoder and hierarchical agglomerative clustering to generate cloud clusters that capture meaningful distinctions among cloud textures, using only raw multispectral imagery as input. Cloud classes are therefore defined based on spectral properties and spatial textures without reliance on location, time/season, derived physical properties, or pre-designated class definitions. We use this approach to generate a unique new cloud dataset, the AI-driven cloud classification atlas (AICCA), which clusters 22 years of ocean images from the Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA’s Aqua and Terra instruments—198 million patches, each roughly 100 km × 100 km (128 × 128 pixels)—into 42 AI-generated cloud classes, a number determined via a newly-developed stability protocol that we use to maximize richness of information while ensuring stable groupings of patches. AICCA thereby translates 801 TB of satellite images into 54.2 GB of class labels and cloud top and optical properties, a reduction by a factor of 15,000. The 42 AICCA classes produce meaningful spatio-temporal and physical distinctions and capture a greater variety of cloud types than do the nine International Satellite Cloud Climatology Project (ISCCP) categories—for example, multiple textures in the stratocumulus decks along the West coasts of North and South America. We conclude that our methodology has explanatory power, capturing regionally unique cloud classes and providing rich but tractable information for global analysis. AICCA delivers the information from multi-spectral images in a compact form, enables data-driven diagnosis of patterns of cloud organization, provides insight into cloud evolution on timescales of hours to decades, and helps democratize climate research by facilitating access to core data.
@Argonne Energy Technology and Security Fellowship is open for applications! Excellent opportunity for those interested in #AI4Science applied to energy, manufacturing, transportation, infrastructure resilience, security, and more!
#postdoc #fellowship
(emphasis txt shade mine)
Happy New Year to all of you dear friends and colleagues. I hope you had a restful and safe break.
If you're like me, you might be feeling recharged and tired simultaneously. While it might be a two coffee day, it's time to jump back into science! 😊 ☕☕
In 2023, we should find ways incentivize and reward data products and software equivalently to journal publications. As we approach an #AI-centric research enterprise, software, services, #openscience, and data products are key components that will automate, connect, and unify distributed effort.
How do we make progress on this?
Started crunching the numbers on #AI/#ML across scientific domains this weekend. Initial screen shows another record year on the way for 2022 - we will release final numbers for 2022 in late Feb. We also potentially have an NLP screening method that will allow us more insight!
Source and data available here: https://github.com/blaiszik/ml_publication_charts
If you are interested in topics of laboratory #data management and microscopy, Northwestern University and Duke University are holding a virtual workshop Dec 8th featuring presentations by Mitra Taheri of The Johns Hopkins University and June Lau of National Institute of Standards and Technology (NIST).
Free sign up here: https://bit.ly/marcn-nu . *Registration closes Dec 5th, so sign up soon!*
