University College London Featured PhD Programmes
University of Kent Featured PhD Programmes
University of Dundee Featured PhD Programmes
University College London Featured PhD Programmes
The Hong Kong Polytechnic University Featured PhD Programmes

How artificial intelligence can improve nuclear physics?

Project Description

In recent years, thanks to the remarkable progress in the domain of artificial intelligence, we have now at our disposal very powerful tools to help our scientific investigation.
Neural Networks or Gaussian Process Emulators (GPE) are now used to reduce the computational cost of complex numerical codes [1], but contrary to a simple interpolation, they are able to grasp the underlying physics using a very reduced set of hypothesis.

The York Nuclear Physics group has started investigating the possible usage of machine learning (ML) methods in 2017 [2] by applying a GPE to simulate the structure of an inner crust of a neutron star. By reducing the typical computational cost by several orders of magnitude, the ML opens up completely new line of research.

At present Dr. Barton and Dr. Pastore are working on the development of a new nuclear mass model based on neural networks (NN) [3]: the preliminary results show that a NN can help reduce the typical discrepancy model/observation by roughly a factor of 3. The resulting masses have now a reasonably low error bar and thus they can be used in astrophysical scenarios as supernovae explosions.

The candidate will work on several ML algorithms using a data-driven approach. During the 3-year project, we will study the possible implementation of several ML methods to various theoretical, experimental, and application problems. This may include improving model accuracy, reducing computational cost, applying Bayesian analysis to improve the accuracy of extrapolations, automating nuclear detector calibrations, and improving the position resolution and performance of nuclear detectors. There will by synergies within the whole of the nuclear physics group since these ML tools will aid our scientific investigations.

[1] Regnier, D., Lasseri, R. D., Ebran, J. P., & Penon, A. (2019). Taming nuclear complexity with a committee of deep neural networks. arXiv preprint arXiv:1910.04132.
[2] Pastore, A., Shelley, M., Baroni, S., & Diget, C. A. (2017). A new statistical method for the structure of the inner crust of neutron stars. Journal of Physics G: Nuclear and Particle Physics, 44(9), 094003.
[3] D. Neil, K. Medler, A. Pastore, C. Barton (in preparation)

Funding Notes

Please note that for PhD projects advertised as “awaiting funding”, we anticipate that the majority of decisions will be made in December 2019.

Email Now

Insert previous message below for editing? 
You haven’t included a message. Providing a specific message means universities will take your enquiry more seriously and helps them provide the information you need.
Why not add a message here
* required field
Send a copy to me for my own records.

Your enquiry has been emailed successfully

FindAPhD. Copyright 2005-2019
All rights reserved.