This is the Winter 2022 lecture schedule with topics and associated older video lectures from previous years. The course this year in Winter 2023 will be very similar order to this, with perhaps some topics in the middle dropped and additional new content nearer to the end of term on Deep Learning advances.

  NOTICE for January 2023!  
  The lecture schedule here is from 2022, use it as a guide only. Most of the essential topics and order will be the same, but some details will be updated in the first couple weeks of class. Mostly, some topics will be dropped or shortened, and a few others will be added, near the end.  

Jump to Week

Weeks 1 - 2

Week View Before Lecture On Date… Lecture Topics Discussed playlist link
1 Join us Live Thursday Jan 6, 2022 Intro Lecture on MS Teams Course Introduction and Logistics  
1 Tuesday, January 11, 2022 What is Data? Data Preprocessing, Cleaning and Normalization Data representations, transformation and normalization (min-max, z-score), Data cleaning methods, dealing with missing data, outlier removal Week 1 Playlist
1 Tuesday, January 11, 2022 Measuring Similarity of Data Comparison measures: dot product, hamming distance, etc. Week 1 Playlist
1 Tuesday, January 11, 2022 Measuring Similarity of Data - Distance Metrics Distance metrics, Minkowski, L1,L2 norms, Mahalanobis distance, cosine similarity Week 1 Playlist
2 Tuesday, January 18, 2022 Training and Validation Methodology - Train, Validate, Test Jacknife estimate, Leave-one-out and K-fold cross validation, Splitting data into Train, Validate, Test sets Week 2 Playlist
2 Tuesday, January 18, 2022 Training and Validation Methodology - Ablation Studies Ablation Studies Week 2 Playlist
2 Tuesday, January 18, 2022 Classification - Introduction Classification Introduction Week 2 Playlist
2 Tuesday, January 18, 2022 Classification - Non Parametric Methods, kNN, Parzen Windows Classification - Non Parametric Methods, kNN, Parzen Windows Week 2 Playlist

*REMINDER: these dates are for last year, 2022, the exact topic and lecture dates for 2023 will be posted later.

Weeks 3 - 4

Week View Before Lecture On Date… Lecture Topics Discussed playlist link
3 Tuesday, January 25, 2022 Performance Evaluation - The Confusion Matrix Confusion Matrices, False Postive Rate, False Negative Rate, Type I and Type II Error, Week 3 Playlist
3 Tuesday, January 25, 2022 Performance Evaluation - ROC and PR Curves ROC curves for parameter tuning, Capacity of a model, Accuracy, precision, f-measure, Precision vs Recall, PR curves Week 3 Playlist
3 Tuesday, January 25, 2022 Parameter Estimation - Bias vs. Variance - Unbiased Estimators - Cross-validation Parameter Estimation Definition, Expectation Operator, Unbiased Estimators, Bernoulli Distribution, Bias vs. Variance Tradeoff, Mean-Squared-Error, Experimental Methodologies to Reduce Bias, Jacknife estimate, Leave-one-out and K-fold cross validation Week 3 Playlist
3 Tuesday, January 25, 2022 The medical test paradox, and redesigning Bayes’ rule Probability Basics, Bayes Theorem Week 3 Playlist
4 Tuesday, February 1, 2022 Parameter Estimation - MLE MLE Week 4 Playlist
4 Tuesday, February 1, 2022 Maximum a Posteriori (MAP) Estimation MAP Week 4 Playlist
4 Tuesday, February 1, 2022 Logistic Regression with MLE and MAP Logistic Regression, MLE, MAP Week 4 Playlist
4 Tuesday, February 1, 2022 Naive Bayes Classifier and Expectation Maximization Expectation Maximization, Naive Bayes Classifier Week 4 Playlist

*REMINDER: these dates are for last year, 2022, the exact topic and lecture dates for 2023 will be posted later.

Weeks 5 - 6

5 Tuesday, February 8, 2022 Lect 5S - Overview - Feature Selection, Extraction and Representation Learning Objective Functions for Feature Selection: Inter-class distance, Mutual Information Week 5 Playlist
5 Tuesday, February 8, 2022 Overview - Feature Selection, Extraction and Representation Learning Feature Extraction vs. Feature Selection Week 5 Playlist
5 Tuesday, February 8, 2022 Principal Component Analysis (PCA) PCA Week 5 Playlist
5 Tuesday, February 8, 2022 Fisher Discriminant Analysis (FDA) and Linear Discriminant Analysis (LDA) LDA, FDA uses within and between class scatter Week 5 Playlist
5 Tuesday, February 8, 2022 t-Stochastic Neighbor Embedding (t-SNE) t-SNE Week 5 Playlist
6 Tuesday, February 15, 2022 Decision Trees - Definition Decision Tree Definitions Week 6 Playlist
6 Tuesday, February 15, 2022 Decision Trees Costs Evaluation Entropy and other Cost Evalutions, Regularizers Week 6 Playlist
6 Tuesday, February 15, 2022 Decision Tree Pruning and Implementations Decision Tree Pruning Week 6 Playlist
6 Tuesday, February 15, 2022 Ensemble Methods Regularization, Boosting, Bagging, Ensemble Tree Methods, Random Forests, Adaboost Week 6 Playlist
6 Tuesday, February 15, 2022 Extremely Randomized Trees, Gradient Boosting and Hoeffding Trees Extremely Randomized Trees, Gradient Tree Boosting (optional topic: Streaming Ensemble Algorithms) Week 6 Playlist
9(!) Tuesday, February 15, 2022 Clustering - Partition Based - kMeans (!) CORRECTION: this topic will be covered in Week 9, after the midterm) k-means – pros/cons, definition, identify clusters, compute new mean, identify new clusters TBD

*REMINDER: these dates are for last year, 2022, the exact topic and lecture dates for 2023 will be posted later.

Week 7 - 8 - Reading Week and Midterm

No new content. Work on Assignment 2 and studying for Midterm.

  • Midterm test on March 1, 2022

Week 9

  • Assignment 2 Due ~March 6, 2022~ March 10, 2022!
Week View Before Lecture On Date… Lecture Topics Discussed Playlist Link
9 Tuesday, March 8, 2022 Unsupervised Learning - Clustering Definitions and Hierarchal Algorithms Distinctions between supervised, unsupervised, and semi-supervised learning. Definitionc of clustering of data, similarity measures, clustering criteria, types of clustering, details of simple clustering models Week 9 Playlist
9 Tuesday, March 8, 2022 Clustering - Partition Based - kMeans k-means – pros/cons, definition, identify clusters, compute new mean, identify new clusters Week 9 Playlist
9 Tuesday, March 8, 2022 Clustering - Density Based - DBScan DBScan – pros/cons, definition, calculation of core/neighbour/noise Week 9 Playlist
9 Tuesday, March 8, 2022 Clustering Evaluation Measures Evaluation of Clustering: Internal and External Measures, Dunn-Index, Separation Index, Rand Index, entropy, Jaccard, F&M Week 9 Playlist
9 Tuesday, March 8, 2022 Anomaly Detection Definitions and Classic Approaches Anomaly Detection Definitions and Classic Approaches Week 9 Playlist
9 Tuesday, March 8, 2022 Anomaly Detection Using Density Estimation density as a measure of anomaly, DBScan algorithm Week 9 Playlist
9 Tuesday, March 8, 2022 Anomaly Detection Isolation Forests and Mondrian Forests binary tree clustering, isolation concept for anomaly detection, Isolation Forests algorithm, Isolation Mondrian Forests algorithm Week 9 Playlist

*REMINDER: these dates are for last year, 2022, the exact topic and lecture dates for 2023 will be posted later.

Weeks 10 - 11

  • Lecture topics : Neural Network and Deep Learning Fundamentals, Making Deep Learning More Effective, Convolutional Neural Networks
  • Assignment 3 Released
Week View Before Lecture On Date… Lecture Topics Discussed Playlist Link
10 Tuesday, March 15, 2022 Introduction to Neural Networks Concepts and History Background of Neural Networks, backpropagation, history of development Week 10 Playlist
10 Tuesday, March 15, 2022 Deep Learning Fundamentals Quicker intro to neural networks, and more info on essentials of deep learning, activitaion functions, optimizers, and the XOR problem, universal approximation theorem, net activation function formulation Week 10 Playlist
10 Tuesday, March 15, 2022 Deep Learning by Gradient Descent Gradient Descent, optimizers Week 10 Playlist
10 Tuesday, March 15, 2022 Effective Deep Learning : Regularization methods Effective Deep Learning : Regularization methods Week 10 Playlist
10 Tuesday, March 15, 2022 Effective Deep Learning: Data Augmentation and Vanishing Gradients Effective Deep Learning: Data Augmentation and Vanishing Gradients Week 10 Playlist
11 Tuesday, March 22, 2022 Convolutional Neural Networks I Convolutional Neural Networks Week 11 Playlist
11 Tuesday, March 22, 2022 Convolutional Neural Networks II Benefits, pros/cons, general design Week 11 Playlist
11 Tuesday, March 22, 2022 Visualizing CNNs Pooling, downsampling, strides, padding Week 11 Playlist
11 Tuesday, March 22, 2022 CNNs - Making Deeper Networks CNNs - Making Deep Networks Week 11 Playlist
11 Tuesday, March 22, 2022 Practical Methodology : Hyper-parameter Tuning Practical Methdology Week 11 Playlist

*REMINDER: these dates are for last year, 2022, the exact topic and lecture dates for 2023 will be posted later.

Weeks 12 -13

  • Lecture topics : Course Review
  • Assignment 3 Due April 2, 2022 (grading done by course staff, not via Kritik)
  • Final Exam: Friday, April 8, 2022 at 7:30pm
Week View Before Lecture On Date… Lecture Topics Discussed Playlist Link
12 Tuesday, March 29, 2022 Inception - Resnet - Densenet Improved deep learning architectures for dealing with very deep networks on very large datasets: Inception, Resnet and Densenet Week 12 Playlist
12 Tuesday, March 29, 2022 Adding Time : Recurrent Neural Networks Adding Time : Recurrent Neural Networks Week 12 Playlist
12 Tuesday, March 29, 2022 Autoencoders Autoencoders Week 12 Playlist
12 Tuesday, March 29, 2022 Denoising Autoencoders Denoising Autoencoders Week 12 Playlist
12 Tuesday, March 29, 2022 Practical Methodology : Hyper-parameter Tuning Practical Methdology (revisit) Week 11 Playlist
13 Tuesday, April 5, 2022 No video yet Transfer Learning (if time permits)  
13 Tuesday, April 5, 2022 No video yet GANs (if time permits)  
13 Tuesday, April 5, 2022 Review Discussion Any topics from Course