Reaching for the limit of stability
(2020) FYSM30 20201Mathematical Physics
Department of Physics
- Abstract
- In this project, calculations for the total binding energy of all even-even nuclei available from experimental data are performed using the HFBTHO program. An artificial neural network is applied to train the information obtained from the HFBTHO calculations and predict the binding energy for the nuclei. The results show
impressive improvements to the HFBTHO program. In recent years, the combination of
scientific research and machine learning algorithms has become a popular and successful practice. Although it is hard to judge whether an algorithm is good enough, especially with the rapid development of computer science, the application of machine learning in nuclear models can be reliable and promising in predicting the nuclear... (More) - In this project, calculations for the total binding energy of all even-even nuclei available from experimental data are performed using the HFBTHO program. An artificial neural network is applied to train the information obtained from the HFBTHO calculations and predict the binding energy for the nuclei. The results show
impressive improvements to the HFBTHO program. In recent years, the combination of
scientific research and machine learning algorithms has become a popular and successful practice. Although it is hard to judge whether an algorithm is good enough, especially with the rapid development of computer science, the application of machine learning in nuclear models can be reliable and promising in predicting the nuclear properties. (Less) - Popular Abstract
- Nowadays, artificial intelligence (AI) is participating more and more in people’s daily life, even though some do not realize that. From speech recognition to graph classification, the algorithms of machine learning (ML) has reached a quite trustworthy level. However, if we apply the ML methods into research in nuclear physics, what will happen?
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9027130
- author
- Cui, Weiyi LU
- supervisor
-
- Gillis Carlsson LU
- Andrea Idini LU
- organization
- course
- FYSM30 20201
- year
- 2020
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- nuclear physics, HFBTHO, DFT, machine learning, artificial neural network
- language
- English
- id
- 9027130
- date added to LUP
- 2020-08-26 08:09:33
- date last changed
- 2020-08-26 08:09:56
@misc{9027130, abstract = {{In this project, calculations for the total binding energy of all even-even nuclei available from experimental data are performed using the HFBTHO program. An artificial neural network is applied to train the information obtained from the HFBTHO calculations and predict the binding energy for the nuclei. The results show impressive improvements to the HFBTHO program. In recent years, the combination of scientific research and machine learning algorithms has become a popular and successful practice. Although it is hard to judge whether an algorithm is good enough, especially with the rapid development of computer science, the application of machine learning in nuclear models can be reliable and promising in predicting the nuclear properties.}}, author = {{Cui, Weiyi}}, language = {{eng}}, note = {{Student Paper}}, title = {{Reaching for the limit of stability}}, year = {{2020}}, }