TA: Ding Liu office hours 9:30-11am Tu/Th

Location: Tu/Th 11:05am-12:20pm, CSB 601

There is no required text, but the following are useful references in addition to the reading material assigned for each class:

- Stuart Russell and Peter Norvig, Artificial Intelligence, A Modern Approach.
- David J. C. MacKay, Information Theory, Inference, and Learning Algorithms.
- Trevor Hastie, Robert Tibshirani, Jerome Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction.
- Dana Ballard, Natural Computation.
- Christopher Bishop, Neural Networks for Pattern Recognition.

On | we will cover | which means that after class you will understand | if before class you have read |
---|---|---|---|

1/19 | Probability Theory | independence, bayes rule | wasserman ch 1, 2, 3 |

1/24 | Information Theory | entropy, kl-distance, coding | mackay ch 2 |

1/26 | Probabilistic Inference | priors: bayesian reasoning, MAP | heckerman |

1/31 | Probabilistic Inference | priors on continuous variables | mackay ch 24 |

2/2 | Minimum Description Length | decision trees | mitchell |

2/7 | Probabilistic Inference | polytree | mackay ch 26 |

2/9 | Expectation Maximization | latent variable clustering | bilmes §1-3 |

2/14 | Independent Component Analysis | source separation | mackay ch 34 |

2/16 | Learning Theory | probably approximately correct | kearns&vazirani ch 1 |

2/21 | Learning Theory | VC dimension | kearns&vazirani ch 2, 3 |

2/23 | Eigenvectors | least squares, PCA | bishop 310-314, appendix E |

2/28 | Nonlinear Dimensionality Reduction | isomap, locally linear embedding | roweis; tenenbaum |

3/2 | Optimization | conjugate gradient | bishop 274-282 |

3/7 | Optimization | Gibbs Sampling, MCMC | mackay ch 29 |

3/9 | Review | ||

3/21 | Midterm | ||

3/23 | Midterm Solutions | ||

3/28 | MCMC, Gibbs | (continued from before midterm) | |

3/30 | Perceptron, Backpropagation | the chain rule | mackay ch 38, 39; bishop 140-148 |

4/4 | Support Vectors | the wolfe dual | burges §3-4 |

4/6 | Support Vectors | the kernel trick | |

4/11 | Hidden Markov Models | forward-backward | bilmes §4 |

4/13 | Reinforcement Learning | q-learning | ballard ch 11 |

4/18 | Reinforcement Learning | partial observability | ballard ch 11 |

4/20 | Games | nash equilibrium | morris 115-131 |

4/25 | Games | learning to co-operate | |

4/27 | Something Fun | ||

5/2 | Review | come to class with questions! |

- Final exam: 35%
- Homeworks: 35%
- Midterm: 25%
- Class participation: 5%

gildea @ cs rochester edu April 11, 2006