CS545: Machine Learning

August

Week	Topic	Material	Reading	Assignments
Week 1: Aug 23, 256	Overview of course. Review of neural networks training and use.	01 Introduction to CS545 02 Searching for Good Weights in a Linear Model	JupyterLab Introduction, watch the video then play with jupyter lab. The Batch from DeepLearning.AI. Yay, Colorado!
Week 2: Aug 30, Sept 1	Thursday lecture cancelled. Please watch pre-recorded lecture in Echo360. Quiz1 and A1 questions. Regression with neural networks.	03 Fitting Simple Models Using Gradient Descent in the Squared Error		Quiz 1 due Wednesday, August 31, 10:00 PM, in Canvas

Week	Topic	Material	Reading	Assignments
Week 3: Sept 6, 8	Introduction to Neural Networks	04 Introduction to Neural Networks 04a Simple Animations	Activation functions in deep learning: A comprehensive survey and benchmark, Neurocomputing, volume 503, 2022, pp. 92-108
Week 4: Sept 13, 15	Python classes. A2.	04b Introduction to Python Classes	Classes Tutorial	A1 Three-Layer Neural Network due Monday, Sept 12th, at 10:00 PM Anderson-Solution-A1
Week 5: Sept 20, 22	Optimizers. Autoencoders.	05 Optimizers 05a Collecting All Weights into One-Dimensional Vector for Use in Optimizers 06 Autoencoders 06a Visualizing Weights	Pandas Cheat Sheet	A2 NeuralNetwork Class due Thursday, Sept 22nd, at 10:00 PM Anderson-A2-Solution
Week 6: Sept 27, 29	A3. Classification	07 Introduction to Classification

Week	Topic	Material	Reading	Assignments
Week 7: Oct 4, 6	Classification. Convolutional neural networks.	08 Classification with Linear Logistic Regression 09 Classification with Nonlinear Logistic Regression Using Neural Networks	The Great AI Reckoning	A3 NeuralNetwork Class Using Optimizers due Thursday, October 6th, at 10:00 PM.
Week 8: Oct 11, 13	Pytorch. Convolutional neural nets	10 JAX 11 Convolutional Neural Networks neuralnetworks_streamlit.tar CNN Backpropagation Notes 12 Introduction to Pytorch	JAX Ecosystem Streamlit Breaking Into AI: Sahar Nasiri on Acing the Data Science Job Interview	A4 Neural Network Classifier due Friday, October 14th, at 10:00 PM. A4 solution available here as A4solution.tar, and here are examples of good solutions.
Week 9: Oct 18, 20	Convolutional Neural Nets in Pytorch. Reinforcement Learnirng	13 Convolutional Neural Networks in Pytorch 15 Introduction to Reinforcement Learning	Toward Next-Generation Artificial Intelligence: Catalyzing the NeuroAI Revolution
Week 10: Oct 25, 27	Reinforcement Learning	16 Reinforcement Learning with Neural Network as Q Function 17 Reinforcement Learning for Two Player Games	Last Week in AI newsletter, with lots of topics for possible semester projects. Pong in a dish	Project Proposal, due Friday, October 28, 10:00 PM

Week	Topic	Material	Reading	Assignments
Week 11: Nov 1, 3	Reinforcement Learning for control dynamical systems. Transfer learning in Reinforcement Learning.	18 Reinforcement Learning to Control a Marble 19 Reinforcement Learning Modular Framework 20 Reinforcement Learning to Control a Marble Variable Goal	Faster Reinforcement Learning After Pretraining Deep Networks to Predict State Dynamics, Increased Reinforcement Learning Performance through Transfer of Representation Learned by State Prediction Model	A5 Convolutional Neural Networks due Friday, November 4th, at 10:00 PM. Here are examples of good solutions.
Week 12: Nov 8, 10	Brain-Computer Interfaces. Linear dimensionality reduction.	21 Linear Dimensionality Reduction with PCA 22 Linear Dimensionality Reduction with Sammon Mapping
Week 13: Nov 15, 17	Recurrent neural networks.	23 Recurrent Neural Networks 24 Recurrent Network Applications		A6 Reinforcement Learning to Control a Robot due Friday, November 18th, at 10:00 PM. Examples of good solutions
Fall Break: Nov 21-25
Week 14: Dec 1	K-means clustering. K-nearest-neighbor classification.	25 K-Means Clustering, K-Nearest-Neighbor Classification

Week	Topic	Material	Reading	Assignments
Week 15: Dec 6, 8	GTA Saira Jabeen summarizes her research. Support Vector Machines. Understanding what a neural net has learned using optimal inputs.	26 Support Vector Machines An Interpretable Model of Climate Change Using Correlative Learning: Slides	An Interpretable Model of Climate Change Using Correlative Learning: Paper Alopex: A Correlation-Based Learning Algorithm for Feedforward and Recurrent Neural Networks: Paper
Dec 12-16	Final Exam Week	No Exams in this course		Project Report, due Monday, December 12th, 10:00 PM. Here is a list of project titles and authors.