AM-IA brings together news, research papers and applications relating to artificial intelligence published around the world into a single feed — and makes them available in your European language at the click of a button.
Scroll to browse · Click to filter
An emulator trained on RADMC-3D spectra enables Bayesian inference on edge-on discs and the study of their ices at the population scale.
Trained on annotations from Galaxy Zoo 3D and images from the DESI Legacy Survey, this publicly available tool is designed for the major surveys to come, which manual catalogues can no longer keep up with.
Trained on a vast dataset of verified calibrations, the model significantly reduces detection errors and could be used in next-generation radio telescopes.
Trained on 850,000 sequences, the model evaluates a route in less than a millisecond with a 5 per cent error rate, and outperforms the best-known solutions to the GTOC9 problem.
By directly fitting the full shape of the spectra rather than a normalised spectrum, a machine learning model outperforms the official DESI pipeline and identifies four additional chemical elements.
Steve Croft describes how machine learning enables Breakthrough Listen to search through archives and real-time data streams – including those from the Vera Rubin Observatory – in the hunt for technosignatures.
To replace Martian models written in old Fortran, researchers are proposing AEGIS, which is capable of simulating ten Martian years and providing gradients for calibration and training.
A simulation-based inference method, which had proved reliable with synthetic data, failed in the case of WASP-39b. The problem lay not with the physics of the simulator but with an overlooked variable: MIRAGE incorporates this variable and reduces the χ²/N from 301 to 0.06.
Thousands of simulated light curves show that an autoregressive detection method would identify at least half of the short-period, Earth-sized planets after just three months of observation.
A normalising flow teaches us how to ‘flatten’ the parameter space of eccentric binaries, enabling us to construct more compact template libraries for the search for gravitational waves.
A study uses predictive validation to test the reliability of representations of celestial objects when switching between different telescopes: the Legacy Survey, HSC, SDSS and JWST.
Stanford researchers anchor the reasoning of an LLM in orbital planning modules: 98% of mission specifications correctly reconstructed with the best models.
Trained on public injection data from the O4 run, the emulator reproduces detection efficiencies and is sometimes more accurate than Monte Carlo sampling.
Without polarization measurements, two researchers reconstruct vector magnetograms close to those of Hinode, and the model is transposed to other instruments.
Millions of synthetic global populations, filtered through Kepler and then compared with actual detections, refine the occurrence rates according to the stars’ mass and metallicity.
By mapping the colours of galactic sources onto the Hubble filter system, this pipeline classifies extragalactic X-ray sources into seven classes with 99.6 per cent accuracy.
Ten learned filters replace the tested acceleration grid one by one. On simulated observations, LeoNet misses an average of 55.5% fewer signals than the evaluated PRESTO configuration.
The first prototype satellite, fitted with a TPU, was launched on 1 October. Google has put a figure on the next phase: 370,000 tonnes to be placed into orbit over ten years, at around $200 per kilo.
Trained on synthetic spectra, this convolutional neural network identifies the signatures of cold atomic gas without modelling the underlying emission, with 96.5 per cent accuracy, at a rate compatible with the future SKA.
Rather than comparing the data with theoretical models, this pipeline detects anomalies. It is all the more effective when the colliding black holes are massive.