arXiv subjects

Edward T. Stevenson

Publications and source records attributed to Edward T. Stevenson.

3 recordsLinked to original sources

arXiv2026

ThousandWorlds: A benchmark for climate emulation of potentially habitable exoplanets

The search for life beyond Earth will depend on detecting faint signatures in the atmospheres of potentially habitable exoplanets. Interpreting those signatures requires understanding the host planet's climate: the same molecule may signal life on one planet and abiotic chemistry on another. Global climate models (GCMs) provide this understanding, but individual runs can require up to millions of core-hours and substantial domain expert time. Machine-learning emulators could remove this bottleneck, but progress has been limited by the absence of a curated, multi-model exoclimate dataset. We introduce ThousandWorlds, an ML-ready benchmark for exoclimate emulation and for the broader regime of low-data, multi-simulator, parameter-to-field regression. The dataset contains approximately 1800 simulations from five GCMs, mapping eight planet parameters to 3D atmospheric fields including temperature, humidity, winds, clouds, and radiation. Three nested subsets define progressively harder challenges: single-simulator regression, multi-simulator regression with complete observations, and multi-simulator regression with structured missingness. We propose two evaluation protocols: one for ranking methods, and one that measures performance relative to the disagreement between GCMs themselves. We evaluate seven baselines spanning simple methods, deep learning, and Gaussian processes. GP-based methods perform best, suggesting that ThousandWorlds exposes a regime where off-the-shelf deep learning does not yet succeed. Data: https://doi.org/10.57967/hf/8695. Code: https://github.com/edstevenson/ThousandWorlds.

cs.LG↗

arXiv2026

Gaussian Process Latent Factor Regression for Low-Data, High-Dimensional Output Problems

In the sciences, regression tasks often require predicting high-dimensional outputs from few training examples. Multi-output Gaussian processes excel in low-data regimes but typically struggle with high-dimensional outputs. Compress-then-predict pipelines such as PCA-GP (principal component analysis plus Gaussian process regression) handle high dimensionality, but rely on bases optimized for reconstruction rather than prediction. To address this gap, we propose a model that represents each output as a linear-Gaussian decoding of a low-dimensional latent state drawn from a Gaussian process prior. By analytically marginalizing the decoder weights, we couple compression and prediction in a single objective that scales to high-dimensional outputs. We refer to this model as Gaussian process latent factor regression (GPLFR). We demonstrate GPLFR by building the first spatially resolved emulator of global climate models for rocky exoplanets.

cs.LG↗

arXiv2024

Searching for planet-driven dust spirals in ALMA visibilities

ALMA (Atacama Large Millimetre/submillimetre Array) observations of the thermal emission from protoplanetary disc dust have revealed a wealth of substructures that could evidence embedded planets, but planet-driven spirals, one of the more compelling lines of evidence, remain relatively rare. Existing works have focused on detecting these spirals using methods that operate in image space. Here, we explore the planet detection capabilities of fitting planet-driven spirals to disc observations directly in visibility space. We test our method on synthetic ALMA observations of planet-containing model discs for a range of disc/observational parameters, finding it significantly outperforms image residuals in identifying spirals in these observations and is able to identify spirals in regions of the parameter space in which no gaps are detected. These tests suggest that a visibility-space fitting approach warrants further investigation and may be able to find planet-driven spirals in observations that have not yet been found with existing approaches. We also test our method on six discs in the Taurus molecular cloud observed with ALMA at 1.33 mm, but find no evidence for planet-driven spirals. We find that the minimum planet masses necessary to drive detectable spirals range from $\approx$ 0.03 to 0.5 $M_{\text{Jup}}$ over orbital radii of 10 to 100 au, with planet masses below these thresholds potentially hiding in such disc observations. Conversely, we suggest that planets $\gtrsim$ 0.5 to 1 $M_{\text{Jup}}$ can likely be ruled out over orbital radii of $\approx$ 20 to 60 au on the grounds that we would have detected them if they were present.

astro-ph.EP↗