Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

$\texttt{Mangrove}$: Learning Galaxy Properties from Merger Trees


Oct 24, 2022
Christian Kragh Jespersen, Miles Cranmer, Peter Melchior, Shirley Ho, Rachel S. Somerville, Austen Gabrielpillai

Add code

* 15 pages, 9 figures, 3 tables, 10 pages of Appendices. Accepted for publication in ApJ 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Accelerated Bayesian SED Modeling using Amortized Neural Posterior Estimation


Mar 14, 2022
ChangHoon Hahn, Peter Melchior

Add code

* 21 pages, 5 figures; submitted to ApJ; code available at https://changhoonhahn.github.io/SEDflow 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Graph Neural Network-based Resource Allocation Strategies for Multi-Object Spectroscopy


Sep 29, 2021
Tianshu Wang, Peter Melchior

Add code

* The GNN code used in this paper is available at https://github.com/tianshu-wang/PFS-GNN-bipartite 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Unsupervised Resource Allocation with Graph Neural Networks


Jun 17, 2021
Miles Cranmer, Peter Melchior, Brian Nord

Add code

* Accepted to PMLR/contributed oral at NeurIPS 2020 Pre-registration Workshop. Code at https://github.com/MilesCranmer/gnn_resource_allocation 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

$\texttt{deep21}$: a Deep Learning Method for 21cm Foreground Removal


Oct 29, 2020
T. Lucas Makinen, Lachlan Lancaster, Francisco Villaescusa-Navarro, Peter Melchior, Shirley Ho, Laurence Perreault-Levasseur, David N. Spergel

Add code

* To be submitted to JCAP. 28 pages, 11 figures 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Hybrid Physical-Deep Learning Model for Astronomical Inverse Problems


Dec 09, 2019
Francois Lanusse, Peter Melchior, Fred Moolekamp

Add code

* 8 pages, accepted submission to the NeurIPS 2019 Machine Learning and the Physical Sciences Workshop 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Block-Simultaneous Direction Method of Multipliers: A proximal primal-dual splitting algorithm for nonconvex problems with multiple constraints


Aug 30, 2017
Fred Moolekamp, Peter Melchior

Add code

* 13 pages, 4 figures 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email