Explainable Machine Learning
Model Interpretability and Visualization
No group and no team. Each person should take one project individually and independently.
Each student should give (a) a proposal, (b) an oral presentation, and (c) a web report.
- Deadlines and reports
Proposal : 12/21, Web page.
Oral Presentation : 01/05, Each person has 10 minutes.
Web Report : 01/10. reading report and implementation results.
- Programming with Python and PyTorch
Guide to Interpretable Machine Learning - Techniques to dispel the black box myth of deep learning. Towards Data Science, 2020.
Interpretability in Machine Learning, Medium, 2020.
Mengnan Du, Ninghao Liu, Xia Hu, "Techniques for Interpretable Machine Learning," Communications of the ACM, Vol. 63 No. 1, Pages 68-77, January 2020.
The great AI debate: Interpretability, Medium, 2019.
J. Yosinski, J. Clune, A. Nguyen, T. Fuchs, H. Lipson, "Understanding Neural Networks Through Deep Visualization," ICML 2015.
Machine Learning Visualization, CNN Visualization
Generate Publication-Ready Plots Using Seaborn Library, 2020/12.
Part-2. Facet, Pair and Joint plots using seaborn
Part-3. Seaborn’s style guide and colour palettes
Part-4. Seaborn plot modifications (legend, tick, and axis labels etc.)
Part-5. Plot saving and miscellaneous
Visdom (GitHub) [PyTorch], Facebook Research
Explainable MNIST classification, 2020.
Papers about interpretable CNN (GitHub)