Michał Bejger

Michał Bejger

Astrophysicist surfing the gravitational waves

(    (   ((  ((( *o )))  ))   )    )

Neural Networks - collection of links

deeplearning.net is a great source of links for various DNN tools. TensorHub collects models, data sets, solutions and other resources. Current tool of choice is the TensorFlow.

List of potentially interesting use of artificial neural networks in astronomy

  • Skynet (ANSII C, MPI): SKYNET: an efficient and robust neural network training tool for machine learning in astronomy https://arxiv.org/abs/1309.0790

  • GUANGTUN BEN ZHU, PhD - tools for cosmology http://www.guangtunbenzhu.com/#!deep-learning/stoom https://github.com/guangtunbenzhu/BGT-Cosmology (BGT - Bimetric Gravity Theory?)

  • Measuring photometric redshifts using galaxy images and Deep Neural Networks. Tree methods and lenet (convolutional neural networks) http://adsabs.harvard.edu/abs/2016A%26C….16…34H

  • Deep Learning Approach for Classifying Galaxies. Convolutional neural networks http://ijarcsse.com/docs/papers/Volume_6/4_April2016/V6I4-0340.pdf

  • Galaxy Zoo Challenge: Classify Galaxy Morphologies from Images. PCA (Principle Component Analysis), SIFT (Scale-Invariant Feature Transform), HOG (Histogram of Oriented Gradients) http://web.stanford.edu/class/cs231a/prev_projects/C231a_final.pdf

  • Rotation-invariant convolutional neural networks for galaxy morphology prediction https://arxiv.org/pdf/1503.07077.pdf http://blog.kaggle.com/2014/04/18/winning-the-galaxy-challenge-with-convnets/

  • Proposal to classify 100 mln galaxies: http://pleiades.ucsc.edu/ams250/projects/Note_Google_proposal_Mar3_2016.pdf A Catalog of Visual-Like Morphologies in the 5 CANDELS Fields Using Deep Learning

  • GPU days 2016, Wigner Center Applications of GPU-accelerated deep neural networks to analyze galaxy spectra http://indico.kfki.hu/event/387/contribution/10

  • Not a DNN, but Generative Artificial NN (GANN) applied for Gaia spectra. On the estimation of stellar parameters with uncertainty prediction from Generative Artificial Neural Networks: application to Gaia RVS simulated spectra

Updated: