Machine learning Archives

A diagram showing how data flows through the transformer architecture implemented here. Data padding and masking are left out for simplicity. The model does implement this functionality, but it is not necessary for our constant-length sequences. -- astro-ph.EP

AI - Data - Apps - Cybernetics, Atmospheres, Climate, Weather, Biogeochemical Cycles & Geobiology, Exoplanets, -moons, -comets, Status Report

Accelerating Radiative Transfer For Planetary Atmospheres by Orders of Magnitude with a Transformer-Based Machine Learning Model

by Keith CowingNovember 7, 2025November 7, 2025

Radiative transfer calculations are essential for modeling planetary atmospheres. However, standard methods are computationally demanding and impose accuracy-speed trade-offs. High computational costs force numerical simplifications in large models (e.g., General […]

Figure reproduced from H. Toru Shay et al. (2025). The detected inventory of TMC–1 molecular species and their column densities is converted into numerical vectors with the VICGAE model. The vectorized strings and the corresponding column densities are then used to train a machine learning regressor. Through learning the relationship between vectorized molecular structures and their column densities, the model is optimized to find the best fit. Once trained, the best-fit model can predict column densities for new molecules and be used to identify potential detection targets in TMC–1. -- The Astrophysical Journal

AI - Data - Apps - Cybernetics, Astrochemistry, Astronomy & Telescopes, Biochemistry & Organic Chemistry, Biosignatures & Paleobiology, Imaging & Spectroscopy, Status Report

Bridging Machine Learning and Spectroscopy: A New Analysis for Astrochemical Target Selection

by Keith CowingOctober 27, 2025October 27, 2025

In the last few years, several approaches have used machine learning to reproduce the inventory and column densities of interstellar molecular sources.

AI - Data - Apps - Cybernetics, Astrobiology (general), Biosignatures & Paleobiology, Reports, Roadmaps, & Plans, Status Report

Foundation Models for Astrobiology: Paper I — Workshop and Overview

by Keith CowingOctober 14, 2025October 14, 2025

Advances in machine learning over the past decade have resulted in a proliferation of algorithmic applications for encoding, characterizing, and acting on complex data that may contain many high dimensional […]

Model prediction on a test sequence from the semi-synthetic TW Hya dataset. The red circle indicates the true position of the injected planet in each frame. In the first frame, the model’s successful prediction is shown with a blue circle and yellow text, demonstrating a position accuracy of a few pixels in a realistic data environment. -- astro-ph.EP

AI - Data - Apps - Cybernetics, Astronomy & Telescopes, Exoplanets, -moons, -comets, Imaging & Spectroscopy, Status Report

Machine Learning Revolution For Exoplanet Direct Imaging Detection: Transformer Architectures

by Keith CowingAugust 21, 2025August 21, 2025

Directly imaging exoplanets is a formidable challenge due to extreme contrast ratios and quasi-static speckle noise, motivating the exploration of advanced post-processing methods.

The radial gas temperature maps of planets from the 3D AFGKM ExoRad with a global temperature, Tglobal of 800K (left), 1600K (center) and 2400K (right) that orbit different host stars (top to bottom: A, G, M). The (Tgas, pgas)-maps gas are shown as equatorial slice plots. -- astro-ph.EP

AI - Data - Apps - Cybernetics, Atmospheres, Climate, Weather, Exoplanets, -moons, -comets, Gas Giants, Status Report

Accelerating Exoplanet Climate Modelling: A Machine Learning Approach To Complement 3D GCM Grid Simulations

by Keith CowingAugust 15, 2025August 15, 2025

With the development of ever-improving telescopes capable of observing exoplanet atmospheres in greater detail and number, there is a growing demand for enhanced 3D climate models to support and help […]

Examples of training reflection spectra for Earth-like exoplanets in the Archean (left), Proterozoic (center), and Modern (right) eras, observed with LUVOIR-B (top row) and Habex/SS (bottom row). Colored lines represent the total radiance 𝐴(𝜆), while points with error bars correspond to 𝐴noisy (𝜆) as defined by Equation 13. -- astro-ph.EP

AI - Data - Apps - Cybernetics, Analog Studies, Astrogeology, Astronomy & Telescopes, Atmospheres, Climate, Weather, Biosignatures & Paleobiology, Habitable Zones, Imaging & Spectroscopy, Press Release

Towards the Habitable Worlds Observatory: 1D CNN Retrieval of Reflection Spectra from Evolving Earth Analogs

by Keith CowingAugust 4, 2025August 4, 2025

However, traditional atmospheric retrieval frameworks are too computationally intensive to explore the high-dimensional parameter spaces such missions will generate.

Multi-layered pipeline for identifying bacterium/archaeon pairs with similar genomic signatures. Layer 1: Five selected non-parametric clustering methods identify clusters of organisms with similar genomic signatures. The clusters containing both bacteria and archaea (green) generate a list of 78 candidate bacterium–archaeon pairs, grouped by these algorithms based on their similar genomic signatures. Layer 2-a:: The candidate pairs from Layer 1 undergo pairwise distance calculations between their FCGRs using four different distance metrics. Only 40 pairs, with the majority of distances below empirically determined thresholds, are retained. Layer 2-b: In parallel to FCGR comparison, a biological analysis is conducted on the output pairs from Layer 1. This includes checking environment labels and examining metadata about their living environments to select pairs isolated from similar types of extreme environments, resulting in 20 pairs. The final output is a list of 15 bacterium/archaeon pairs (comprising 16 unique genera and 20 unique species) that passed all filtering layers. These pairs can confidently be proposed as maximally divergent microbes that share similar genomic signatures associated with their living environments. -- biorxiv.org

AI - Data - Apps - Cybernetics, Biochemistry & Organic Chemistry, Biophysics, Biosignatures & Paleobiology, Extremeophiles and Extreme Environments, Genomics, Proteomics, Bioinformatics, Microbiology & Virology, Nomenclature, Systematics, Natural History, Status Report

Life At The Extremes: Maximally Divergent Microbes With Similar Genomic Signatures Linked To Extreme Environments

by Keith CowingJune 30, 2025June 28, 2025

Extreme environments impose strong mutation and selection pressures that drive distinctive, yet understudied, genomic adaptations in extremophiles.

AI - Data - Apps - Cybernetics, Astrogeology, Astronomy & Telescopes, Biosignatures & Paleobiology, Enceladus, Europa, Exoplanets, -moons, -comets, Habitable Zones, Imaging & Spectroscopy, Mapping, Geodesy, Cartography, Bathymetry, Origin & Evolution of Life, Status Report, Water/Hycean Worlds & Oceanography

Interpretable Machine Learning Biosignature Detection From Ocean Worlds Analogue CO2 Isotopologue Data

by Keith CowingMay 30, 2025May 26, 2025

Future missions to icy ocean worlds (OW) such as Europa and Enceladus will evaluate the habitability and potential for biosignatures on these worlds.

AI - Data - Apps - Cybernetics, Astrobiology (general), Citizen Science / Open Science, Genomics, Proteomics, Bioinformatics, NASA Astrobiology Program, Reports, Roadmaps, & Plans, Status Report

The Astrobiology Data Ecosystem, Open Science, And The AI Era – NASA-DARES 2025 White Paper

by Keith CowingMay 24, 2025January 14, 2026

DARES Alignment · Primary Topic: Identify Emerging Themes and Technologies· Secondary Topic(s): Review Recent Advancements, Strengthen Community

AI - Data - Apps - Cybernetics, Astrobiology (general), NASA Astrobiology Program, Press Release, Reports, Roadmaps, & Plans

Foundation Models For Astrobiology – NASA-DARES 2025 White Paper

by Keith CowingMay 22, 2025January 14, 2026

A white paper for the 2025 NASA Decadal Astrobiology Research and Exploration Strategy (DARES)

Machine learning

Accelerating Radiative Transfer For Planetary Atmospheres by Orders of Magnitude with a Transformer-Based Machine Learning Model

Bridging Machine Learning and Spectroscopy: A New Analysis for Astrochemical Target Selection

Foundation Models for Astrobiology: Paper I — Workshop and Overview

Machine Learning Revolution For Exoplanet Direct Imaging Detection: Transformer Architectures

Accelerating Exoplanet Climate Modelling: A Machine Learning Approach To Complement 3D GCM Grid Simulations

Towards the Habitable Worlds Observatory: 1D CNN Retrieval of Reflection Spectra from Evolving Earth Analogs

Life At The Extremes: Maximally Divergent Microbes With Similar Genomic Signatures Linked To Extreme Environments

Interpretable Machine Learning Biosignature Detection From Ocean Worlds Analogue CO2 Isotopologue Data

The Astrobiology Data Ecosystem, Open Science, And The AI Era – NASA-DARES 2025 White Paper

Foundation Models For Astrobiology – NASA-DARES 2025 White Paper

NASA Spaceline Current Awareness List #1,208 17 July 2026 (Space Life Science Research Results)

Tidal Evolution Of Packed Moon Systems Around An Earth-mass Planet

Deep-sea Creatures’ Epic Migrations Between Hydrothermal Vents

Efficient Interstellar Grain Growth From High Sticking Coefficients On Amorphous Carbon Dust

A Pressure-Supported Pneumatic Architecture for Robust Deep Drilling on Mars: Performance Modeling, Sizing, and Mission Integration.