EECS282 Advanced Topics in Machine Learning (Fall semester 2021)

Instructor

Miguel Á. Carreira-Perpiñán
Professor
Electrical Engineering and Computer Science
School of Engineering
University of California, Merced
mcarreira-perpinan-[at]-ucmerced.edu; 209-2284545
Office: 217, Science & Engineering Building 2

Office hours: by appointment in SE2-217 or Zoom.

Lectures: Mondays/Wednesdays 12-1:15pm (COB2-272).

Lab class: Fridays 4:30-7:20pm (Linux Lab, SE1-100).

Course web page: http://faculty.ucmerced.edu/mcarreira-perpinan/teaching/EECS282

Course description

The course reviews advanced topics in machine learning. Machine learning is the study of models and algorithms that learn information from data. Machine learning ideas underlie many algorithms in computer vision, speech processing, bioinformatics, robotics, computer graphics and other areas. The 2021 edition of the course will focus on machine learning model interpretability, broadly understood.

Prerequisites: prior knowledge about machine learning at the level of an introductory course (e.g. having taken CSE176 at UC Merced, or equivalent), or instructor approval.

Syllabus

Textbook

There is no required textbook. Selected readings will appear in this web page in due course. The following are some general books about machine learning:

• C. Bishop: Pattern Recognition and Machine Learning. Springer, 2006. Companion site.
• T. J. Hastie, R. J. Tibshirani and J. H. Friedman: The Elements of Statistical Learning: Data Mining, Inference and Prediction, 2nd ed. Springer, 2009. This book is accessible online from within UC Merced. Companion site.
• K. P. Murphy: Machine Learning: A Probabilistic Perspective. MIT Press, 2012. Companion site.
• S. Russell and P. Norvig: Artificial Intelligence: A Modern Approach, 4th ed. Pearson, 2021. Companion site.
• Carreira-Perpiñán, M. Á. (2016): CSE176 Introduction to Machine Learning: Lecture notes. University of California, Merced, 2015-2016.

Practical project

TBD

Schedule of presentations

• Aug 30: introduction to interpretable models in machine learning (presenter: Miguel Á. Carreira-Perpiñán)
• Sep 1: overview of machine learning (presenter: Miguel Á. Carreira-Perpiñán)
• Sep 3: overview of deep neural nets 1: convolutional neural nets (presenter: Yerlan Idelbayev)
• Sep 8: overview of deep neural nets 2: training techniques (presenter: Yerlan Idelbayev)
• Sep 10: overview of deep neural nets 3: transformers (presenter: Yerlan Idelbayev and Arman Zharmagambetov)
• Sep 13: overview of deep neural nets 4: NLP (presenter: Arman Zharmagambetov)
• Sep 15: word embeddings 1: TFIDF, LSA (presenter: Suryabhan Singh Hada)
• Sep 17: word embeddings 2: word2vec, GloVe, fastText (presenter: Arman Zharmagambetov)
• Sep 20: word embeddings 3: demo (presenter: Suryabhan Singh Hada)
• Sep 24: interpretability of attention (presenter: Hoa Nguyen)
• Sep 27: overview of splines (presenter: Magzhan Gabidolla)
• Sep 29: generalised additive models 1 (presenter: Magzhan Gabidolla)
• Oct 1: generalised additive models 2 (presenter: Magzhan Gabidolla)
• Oct 4: counterfactual explanations with decision trees (presenter: Suryabhan Singh Hada)
• Oct 6: understanding neural nets: the inverse set of a neuron (presenter: Suryabhan Singh Hada)
• Oct 8: adversarial examples (presenter: Aditya Petety and Suryabhan Singh Hada)
• Oct 11: nonlinear embeddings for dimensionality reduction (presenter: Miguel Á. Carreira-Perpiñán)
• Oct 13: interpretable clustering 1 (presenter: Magzhan Gabidolla)
• Oct 15: interpretable clustering 2 (presenter: Magzhan Gabidolla)
• Oct 18: interpretable clustering 3 & interpretable dimensionality reduction 1 (presenter: Magzhan Gabidolla, Arman Zharmagambetov)
• Oct 20: sparse PCA and interpretable dimensionality reduction 2 (presenter: Arman Zharmagambetov)
• Oct 25: ensembles (presenter: Magzhan Gabidolla, Arman Zharmagambetov)
• Oct 27: scoring models and rule lists 1 (presenter: Yerlan Idelbayev)
• Nov 1: scoring models and rule lists 2 (presenter: Yerlan Idelbayev)
• Nov 17: scoring models and rule lists 3 (presenter: Yerlan Idelbayev)
• Nov 30: interpretable dimensionality reduction 3 (presenter: Arman Zharmagambetov)
• Dec 2: scoring models and rule lists 4 (presenter: Yerlan Idelbayev)
• Dec 3: feature importance 1 (presenter: Suryabhan Singh Hada)
• Dec 7: feature importance 2 (presenter: Suryabhan Singh Hada)
• Dec 8: feature importance 3 (presenter: Suryabhan Singh Hada)
• Dec 9: model interpretability in practice (presenter: Magzhan Gabidolla)

UC Merced | EECS | MACP's Home Page