Analysis of Galactic Spectra using Noise-Aware Learning Algorithms

All Papers

We introduce a novel learning algorithm for noise elimination. Our algorithm is based on the re-measurement idea for the correction of erroneous observations and is able to discriminate between noisy and noiseless observations by using kernel methods. We apply our noise-aware algorithms to the prediction of stellar population parameters, a challenging astronomical problem. Experimental results adding noise and useful anomalies to the data show that our algorithm provides a significant reduction in error, without having to eliminate any observation from the original dataset.