Project information

(Under construction)

| Jan 16 | Supervised learning paradigm; decision trees | | CIML Chap. 2 | | | Jan 18 | Finish decision trees; limits of learning; practical issues in supervised learning | | CIML Chap. 3 | HW0 due | | Jan 23 | Hyperparameter selection; cross-validation; nearest neighbors | | CIML Chap. 4 | HW1 | | Jan 25 | Labor day holiday | | | | | Jan 30 | Linear classifiers; Perceptron | | CIML 5.1-5.6 | | | Feb 1 | Practical considerations: features; performance metric beyond error; ROC curves | | CIML 5.7, 7.1-7.3 | | | Feb 6 | Precision-recall curves; confidence intervals and hypothesis testing | | CIML 7.4-7.6 | | | Feb 8 | Finish hypothesis testing; Linear models for regression | | CIML 7.7 | | | Feb 13 | Introduction to convex optimization; Linear models for classification | | CIML 11.1-11.2 | | | Feb 15 | SVM; stochastic gradient descent; dual formulations | | CIML 11.4, 11.5 | | | Feb 20 | Finish SVM dual formulation; geometric interpretation; begin kernel methods | | CIML 3.4, 11.3 | | | Feb 22 | Finish kernel methods: kernel Perceptron; kernel ridge regression; begin unsupervised learning: clustering and k-means | | CIML Chap 15 | HW2 | | Feb 27 | Class canceled - please review your HW1 & soln guide | | My old notes on SVM and kernels https://zcc1307.github.io/courses/csc665fa19/notes/svm.pdf; | | | Feb 29 | Finish clustering; Principal component analysis| | CIML 9.1-9.3 | | | Mar 5 | Finish PCA; begin probabilistic graphical models | | CIML 9.4-9.7 | | | Mar 7 | Finish Naive Bayes; Midterm review | | CIML 16.1-16.2 | | | Mar 12 | Midterm exam | | | | | Mar 14 | Expectation-maximization and Gaussian mixture models | | CIML 16.3 | | | Mar 19 | Finish GMM; A closer look at Bayes Nets and Conditional Independence | | Jason Pacheco’s PGM note https://www2.cs.arizona.edu/~pachecoj/courses/csc535_fall20/lectures/pgms.pdf | Project proposal due | | Mar 21 | d-separation; Begin HMMs: the forward algorithm | | Jason Pacheco’s Dynamical Systems note https://www2.cs.arizona.edu/~pachecoj/courses/csc535_fall20/lectures/dynamicalsys.pdf | | | Mar 26 | Finish HMMs: Viterbi algorithm for decoding and EM algorithm for learning | | CIML 10.1-10.2 | | | Mar 28 | Neural Networks; begin backpropagation | | CIML 10.3-10.6 | | | Apr 2 | Finish backpropagation; tips for training neural networks | | Berkeley CS 285 PyTorch tutorial - see links in slides | | | Apr 4 | Convolutional neural networks | | Stanford CS 231n notes on CNNs https://cs231n.github.io/convolutional-networks/ | | | Apr 9 | CNNs cont’d; begin generative models: autoencoders | | | | | Apr 11 | Generative models: VAEs | | Jaan Altosaar’s VAE tutorial https://jaan.io/what-is-variational-autoencoder-vae-tutorial/ | | | Apr 16 | Generative models: GANs | | The original GAN paper https://arxiv.org/pdf/1406.2661.pdf | HW4 | | Apr 18 | Reinforcement learning: policy evaluation and planning in MDPs | | | | | Apr 23 | Reinforcement learning: learning in MDPs | | Andrej Karpathy’s blogpost on Deep RL http://karpathy.github.io/2016/05/31/rl/ | | | Apr 25 | Learning Theory: the PAC and agnostic PAC framework | | | | | Apr 30 | Learning theory: uniform convergence and structural risk minimization; Final review | | ``Deep double descent’’ paper by Nakkiran et al https://openreview.net/forum?id=B1g5sA4twr | | )