**COSC 4550 / COSC 5550**

M/W/F 1:10-2:00pm

Room: Engineering 3076

Professor Nick Cheney

ncheney@uwyo.edu

Office hours: W 2-3pm, F 12-1pm (Engineering 4081B)

TA: Joost Huizinga

jhuizing@uwyo.edu

TA office hours: Tu 2-5pm, Th 1:30-3:30pm (Engineering 4086)

Syllabus (pdf)

Assigned readings should be completed * before* the class to which they are listed. Having a primer on the topics before going into the lectures will allow you to be more engaged with the subtler points in lectures and be able to ask more questions about the parts of the readings that you did not fully grasp. You are fully capable of reading the material in the textbook on your own -- so the lectures are not simply a time for me to show you that material again, but rather are a time for me to compliment the readings you do with real time help and feedback for your questions, to emphasize the parts of the readings that I find most relevant to our class, and to solidify the concepts in the readings with additional examples and hands-on exercises. In this spirit, all of the material from the readings will

(Note: "RN 1-1.5" translates to: In the Russell and Norvig textbook, please read from the start of chapter 1 to the end of section 1.5)

Day | Date | Topic | Assigned Reading | Assignments | Slides |
---|---|---|---|---|---|

W | 8/30 | Course Logistics and Motivation | -- | ||

F | 9/1 | What is AI? Brief History of AI | RN 1-1.5 (30 pages) | ||

M | 9/4 | NO CLASS: Labor Day | -- | -- | -- |

W | 9/6 | The Structure of Intelligent Agents | RN 2-2.5 (25 pages) | ||

F | 9/8 | Problem Solving Agents and | RN 3-3.4.2 (21 pages) | ||

M | 9/11 | Problem Solving with Basic Search Strategies | RN 3.4.3-3.7 (25 pages) | AIC 1 Due 9/10 | |

W | 9/13 | Heuristic Search, Local Search, and Optimization | RN 4-4.1 (9 pages) | ||

F | 9/15 | Optimization Algorithms and Search in Continuous Spaces | RN 4.2 (4 pages) | ||

M | 9/18 | Optimal Decisions in Game Playing | RN 5-5.3 (10 pages) | AIC 2 Due 9/17 | |

W | 9/20 | Refresher on Probabilty (guest lecture) | RN 13-13.4 (15 pages) | ||

F | 9/22 | Evaluation Functions and Games of Chance | RN 5.4-5.9 (not 5.6.1) (18 pages) | ||

M | 9/25 | Bayes Rule | RN 13.5-13.7 (8 pages) | AIC 3 Due 9/24 | |

W | 9/27 | The Semantics of Bayesian Networks | RN 14-14.3 (12 pages) | ||

F | 9/29 | Inference in Bayesian Networks | RN 14.4-14.5 (17 pages) | ||

M | 10/2 | Probabilistic Reasoning Over Time | RN 15-15.2 (12 pages) | ||

W | 10/4 | Hidden Markov Models and Kalman Filters | RN 15.3-15.4 (11 pages) | ||

F | 10/6 | Dynamic Bayesian Networks, Sequential Decision Making | RN 15.5-15.7, 17.1 (18 pages) | ||

M | 10/9 | Value Iteration and Policy Iteration | RN 17.2-17.4 (14 pages) | AIC 4 Due 10/8 | |

W | 10/11 | Decision Trees | RN 18-18.3 (14 pages) | ||

F | 10/13 | Regression and Classification | RN 18.6 (10 pages) | pdf code | |

M | 10/16 | Machine Learning Theory | RN 18.4-18.5 (9 pages) | pdf code | |

W | 10/18 | Artificial Neural Networks | RN 18.7-18.7.3 (5 pages) | ||

F | 10/20 | Learning in Multilayer Networks | RN 18.7.4-18.7.5 (4 pages) | ||

M | 10/23 | Learning in Multilayer Networks Part 2 | RN 18.8 (7 pages) | AIC 5 Due 10/22 | |

W | 10/25 | Non-Parametric Models and Support Vector Machines | RN 18.9-18.12 (14 pages) | ||

F | 10/27 | Deep Learning for Image Classification | Deep Learning (LeCun Bengio Hinton, Nature 2015) (7 pages) | Project Outline Due 10/27 | |

M | 10/30 | NO CLASS: Advising Week | -- | -- | -- |

W | 11/1 | NO CLASS: Advising Week | -- | -- | -- |

F | 11/3 | NO CLASS: Advising Week | -- | -- | -- |

M | 11/6 | Deep Learning Architectures | Imagenet Classification With Deep Convolutional Neural Networks (Krizhevsky Sutskever Hinton, NIPS 2012) (8 pages) | ||

W | 11/8 | Deep Learning Tricks | Dropout: A Simple Way to Prevent Neural Networks from Overfitting (Srivastava et al., JMLR 2014) (24 pages) | ||

F | 11/10 | Deep Reinforcement Learning | Human-level control through deep reinforcement learning (Mnih et al., Nature 2015) (4 pages) | ||

M | 11/13 | Deep Reinforcement Learning Part 2 | End to End Training of Deep Visuomotor Policies (Levine et al., JMLR_2016) (27 pages) | ||

W | 11/15 | Unsupervised Learning and Generative Adversarial Networks | Unsupervised representation learning with deep convolutional generative adversarial networks (Radford et al., ICLR 2016) (10 pages) | ||

F | 11/17 | Unsupervised Learning Part 2 | -- | Project Check-in Due 11/17 | |

M | 11/20 | Exam Day! | Sample Exam | -- | -- |

W | 11/22 | NO CLASS: Thanksgiving Break | -- | -- | -- |

F | 11/24 | NO CLASS: Thanksgiving Break | -- | -- | -- |

M | 11/27 | Recurrent Neural Networks and Long Short Term Memory | LSTM: a search space odyssey (Greff et al., IEEE Transactions on Neural Networks and Learning Systems 2015) (8 pages) | ||

W | 11/29 | Preprocessing and Data Pipelines | Data Preprocessing for Supervised Leaning (Kotsiantis et al., IJCIE 2007) (5 pages) | ||

F | 12/1 | Machine Learning Examples in Python | code | ||

M | 12/4 | Embodied Intelligence | Morphological change in machines accelerates the evolution of robust behavior (Bongard, PNAS 2011) (6 pages) | ||

W | 12/6 | The Ethics of Artificial Intelligence | RN 26-26.4 (20 pages) | ||

F | 12/8 | Final Project Presentations | -- | Project Video Due 12/7 | |

M | 12/11 | Final Project Presentations | -- | ||

M | 12/18 | Final Project Presentations | Time: 1:15-3:15pm |

The AI Challenges are hands-on opportunities to demonstrate your knowledge of the concepts covered in class as well as your ability to think creatively and code-up effective and cool programs! The assigned algorithm/agent implementations are meant to test your grasp of these concepts, and thus are to be done independently. This means to code sharing between students, and no usage of existing AI or machine learning packages -- though imports of basic data structures (e.g. NumPy arrays) are completely acceptable.

(Note: all assignments are due by 11:59pm on the day listed below)

Day | Due Date | Assignment | Topics Covered | Link |
---|---|---|---|---|

M | 9/11 | AI Challenge #1 | Building Your First Intelligent Agent | zip |

Su | 9/17 | AI Challenge #2 | Basic Search Strategies | zip |

Su | 9/24 | AI Challenge #3 | Advanced Search Strategies | zip |

Su | 10/8 | AI Challenge #4 | Game Playing Agents | zip |

Su | 10/22 | AI Challenge #5 | Optimal Behavior in Uncertain Environments | zip |

Course projects are a way to get your hands dirty and work on real problems that you care about, while showing off what you've learned this semester and becoming a practicing AI researcher! Projects are meant to be undertaken individually (but please ask if you are interested in a joint project between multiple people -- or between multiple classes). You will be asked to impelment a variation of a method we've shown in class and/or use AI to solve a new problem we haven't talked about in class. At the end of the semester, you'll create a video to share with your class (and the world!) and talk about what you've made. A couple check-in along the way are meant to keep you on track. I've listed the dates for these check in's below, and they should be emailed to me by 11:59pm on the day listed.

Day | Due Date | Deliverable | Description |
---|---|---|---|

F | 10/27 | Project Proposal/Outline | |

F | 11/17 | Mid-project Check-in | |

Th | 12/7 | Final Project Video Due! |