SURF: Announcements of Opportunity
Below are Announcements of Opportunity posted by Caltech faculty and JPL technical staff for the SURF program. Additional AOs for the Amgen Scholars program can be found here.
Specific GROWTH projects being offerred for summer 2018 can be found here.
Each AO indicates whether or not it is open to non-Caltech students. If an AO is NOT open to non-Caltech students, please DO NOT contact the mentor.
Announcements of Opportunity are posted as they are received. Please check back regularly for new AO submissions! Remember: This is just one way that you can go about identifying a suitable project and/or mentor.
Announcements for external summer programs are listed here.
Students pursuing opportunities at JPL must be
U.S. citizens or U.S. permanent residents.
|Project:||Classifying Sparse Light-Curves Using Deep Learning|
|Disciplines:||Data Science, Astronomy|
Sr. Research Scientist, (PMA),
|Mentor URL:||http://www.astro.caltech.edu/~aam (opens in new window)|
Astronomical surveys have hundreds of millions of light curves (time series), a rich dataset to carry out many data science experiments. One challenge is that the observations are far from continuous, or even regular. As a result most traditional methods can not be applied. This is true for past and present surveys like CRTS/CRTS II, PTF/iPTF/ZTF, Gaia, Pan-STARRS, and will be true for future surveys like LSST (to mention just a few and only at optical wavelengths). Additionally, many times one needs rapid classification to follow objects that are fading fast, and that means one has access to a light curve that has only a few points. Identifying the signature of a source under such circumstances requires novel methods.
The path to identifying signatures of different types of objects from few points is to start with good priors, and extending classification schemes to classes with fewer objects and points in their light curves. Recently we applied convolutional neural networks (CNN) to images formed from light curves using the dmdt method for periodic stars. The proposed work will involve extending the method to other classes including stochastic, irregular, and transient sources. We will also compare with other methods applied to time-series data like the Long Short Term Memory (LSTM). We will mainly make use of light curves from CRTS/PTF but will also use public datasets. The aim is to apply the classifications to ZTF that is currently going through commissioning.
Deep Learning: http://deeplearning.net/tutorial/
dmdt method: https://arxiv.org/pdf/1709.06257.pdf
Proficiency in python, conversant with statistics basics, deep learning, using GPUs and AWS, knowledge about linux/unix, git and related software engineering techniques. Basic astronomy knowledge will be a plus.
This AO can be done under the following programs:
<< Prev Record 51 of 151 Next >> Back To List