[AI6103] Deep Learning & Applications
====== I M P O R T A N T ======
The content in this repository should exclusively be utilized in sharing solutions for projects, communicating ideas for related problems, and references to similar assignments. If you are a student facing an assignment with the same or similar topics, you can use this repository as a reference, while the final report should include the citations of the repository. If you submit an assignment without proper acknowledgment after referring to this repository, you may be considered PLAGIARISM by your instructor, and the author will not pay ANY responsibility for this. Please refer to your teacher's and your school's instructions for the determination of academic integrity.
Moreover, if you are taking the AI6103 course, do not be stupid. You can utilize the materials here as a reference to construct your own assignment and reflect the citation to this repository in the final report. If you copy the code without citing it, you have violated NTU's academic integrity and are involved in plagiarism.
Please refer to the following links for NTU's determination of academic integrity and plagiarism:
https://ts.ntu.edu.sg/sites/intranet/dept/tlpd/ai/Pages/NTU-Academic-Integrity-Policy.aspx
https://ts.ntu.edu.sg/sites/intranet/dept/tlpd/ai/Pages/default.aspx
If you think the professor is easy to fool, think again.
You know who you are.
====== D I S C L A I M E R ======
This repository should only be used for reasonable academic discussions. I, the owner of this repository, never and will never ALLOWING another student to copy this assignment as their own. In such circumstances, I do not violate NTU's statement on academic integrity as of the time this repository is open (22/01/2022). I am not responsible for any future plagiarism using the content of this repository.
====== I N T R O D U C T I O N ======
[AI6103] Deep Learning & Applications is a core course of Master of Science in Artificial Intelligence Graduate Programme (MSAI), School of Computer Science and Engineering (SCSE), Nanyang Technological University (NTU), Singapore. The repository corresponds to the AI6103 of Semester 1, AY2021-2022, starting from 08/2021. The instructor of this course is Prof. Li Boyang, Albert.
The projects of this course consist of one individual Homework and one group Project, which including an oral presentation. The topic of the assignment are shown below, and the specific score is not provided to us. Since multiple people complete the group work, I do not have the right to disclose the report and others' codes individually so that the relevant parts will be hidden.
| Type | Topic | Grade |
|---|---|---|
| Quiz | N.A. | N.A. / 20.0 |
| Homeword | Effects of hyperparameters on network training and generalization | N.A. / 30.0 |
| Project | Topic of your choice* | N.A. / 40.0 |
| Oral Presentations | With the project | N.A. / 10.0 |
*Topics for Project:
- Take a conference paper (AAAI, IJCAI, ICML, NeurIPS, ICLR, CVPR, ICCV NAACL, ACL, etc) from 2015. Reimplement the algorithm in that paper. [No existing source code is allowed]
- Take a conference paper in 2019. Develop/implement an improvement on top of its existing code base.
- Take a dataset (e.g., CIFAR-10, Sentiment Analysis, etc.) and a popular network (e.g., ResNet-18, Transformer), and experiment with 4-5 different regularization techniques. Analyze the differences in the result created by these techniques. You may analyze the network using interpretation techniques that we discuss.
====== A C K N O W L E D G E M E N T ======
All of above projects are designed by Prof. Li Boyang, Albert.