papers

This repository maintains a list of papers that our lab looks into.

Content

1. Hoang Le's work are here

2. Huiying's work are here

3. Jiefei Liu's work are here

4. Erick's work are here

5. Selective papers marked by Huiping here

6. Physics guided NN models

7. Steps to set up environment for Federated Learning here

Physics guided NN models

5. Gibson J, Hire A, Hennig RG. Data-augmentation for graph neural network learning of the relaxed energies of unrelaxed structures. npj Computational Materials. 2022 Sep 30;8(1):211.

6. Physics-InformedNeuralNetworks:ADeepLearning FrameworkforSolvingForwardandInverseProblems InvolvingNonlinearPartialDifferentialEquations. Journal of Computational Physics, Volume 378, 1 February 2019, Pages 686-707 (https://www.sciencedirect.com/science/article/abs/pii/S0021999118307125)

7. Physics-Guided Machine Learning for Scientific Discovery: An Application in Simulating Lake Temperature Profiles. Machine Learning (cs.LG), 2020. paper

   Xiaowei Jia, Jared Willard, Anuj Karpatne, Jordan S Read, Jacob A Zwart, Michael Steinbach, Vipin Kumar

8. Physics Guided RNNs for Modeling Dynamical Systems: A Case Study in Simulating Lake Temperature Profiles. Computational Physics (physics.comp-ph), 2019. paper

   Xiaowei Jia, Jared Willard, Anuj Karpatne, Jordan Read, Jacob Zwart, Michael Steinbach, Vipin Kumar

9. Label-Free Supervision of Neural Networks with Physics and Domain Knowledge. AAAI, 2017. paper

   Russell Stewart, Stefano Ermon

Steps to set up the Kaiju environment

Trung's set up on (i.e Kaiju, Shamir, Hidam, Epscor23, etc)

1. Install Anaconda If you have NOT installed Anaconda in your home directory (~/<your_cs_username>) yet, then do this

rm -f ~/anaconda3

mkdir /home/student_no_backup/<your_cs_username>/anaconda3

ln -s /home/student_no_backup/<your_cs_username> ~/anaconda3

These commands create a symlink between your anaconda in your home directory to anaconda in your work directory because your home directory's capacity is not big enough to store the anaconda. In case you did install Anaconeda in your home directory (~/<your_cs_username>), do this

rm -rf /home/student_no_backup/<your_cs_username>/anaconda3

mv ~/anaconda3 /home/student_no_backup/<your_cs_username>/

ln -s /home/student_no_backup/<your_cs_username>/anaconda3 ~/anaconda3

After that you can install Anaconda by following these steps: https://kengchichang.com/post/conda-linux/

2. After installing Anaconda You can put your working projects inside this directory: /home/student_no_backup/<your_cs_username> and create a new conda environment for each project. Enviroments can be used accross servers (Kaiju, Shamir,...)

3. Using Screen Have you ever faced the situation where you perform a long-running task on a remote machine, and suddenly your connection drops, the SSH session is terminated, and your work is lost. Well, it has happened to all of us at some point, hasn’t it? Luckily, there is a utility called screen that allows us to resume the sessions. Screen is preinstalled on most linux machine.

Starting Named Session

screen -S "session-name"

It’s always a good idea to choose a descriptive session name.

Detach from Linux Screen Session

Ctrl+a d

The program running in the screen session will continue to run after you detach from the session.

Reattach to a Linux Screen

screen -r "session-name"

To find the session ID list the current running screen sessions with

screen -ls