Machine Learning for Pulsar Classification

Developed an artificial neural network, based on “Inception v3”, by using transfer learning with Tensorflow, for post-processing data from Arecibo telescope and GBT to detect neutron stars. Achived 19 times increase in efficiency.

Files

Metadata

Work Title Machine Learning for Pulsar Classification
Access
Open Access
Creators
  1. Suprun, Dmytro
  2. Ann Schmiedekamp, Ph.D
  3. Carl Schmiedekamp, Ph.D
Keyword
  1. CNN
  2. Dmytro Suprun
  3. tensorflow
  4. AI
  5. Abington
  6. machine learning
  7. ACURA
  8. convolutional neural network
License GNU General Public License (GPLv3)
Work Type Poster
Deposited May 14, 2017

Versions

Analytics

Collections

This resource is currently not in any collection.

Work History

Version 1
published

  • Created
  • Added wdb78td89z_version1_Final_draft.pdf
  • Added Creator Suprun, Dmytro
  • Added Creator Ann Schmiedekamp, Ph.D
  • Added Creator Carl Schmiedekamp, Ph.D
  • Published