|
課程名稱 |
預測、學習、與賽局
Prediction, Learning, and Games |
|
開課學期 |
112-2 |
|
授課對象 |
電機資訊學院 資訊工程學系 |
|
授課教師 |
李彥寰 |
|
課號 |
CSIE5002 |
|
課程識別碼 |
922 U4550 |
|
班次 |
|
|
學分 |
3.0 |
|
全/半年 |
半年 |
|
必/選修 |
選修 |
|
上課時間 |
星期四7,8,9(14:20~17:20) |
|
上課地點 |
資110 |
|
備註 |
Theory course, requiring math maturity.
總人數上限：20人 |
|
|
|
|
課程簡介影片 |
|
|
核心能力關聯 |
核心能力與課程規劃關聯圖 |
|
課程大綱
|
|
為確保您我的權利,請尊重智慧財產權及不得非法影印
|
|
課程概述 |
This is a *theory* course. There will not be any programming assignment. The students will have to read and write mathematical proofs.
Will it rain tomorrow? Will the stock price go down tomorrow? How likely will the sun rise tomorrow? Such problems can be reduced to a very basic problem: Given a sequence of bits, on which there is not any probabilistic model, how likely will the next bit be 1?
In this course, we will study this problem and its extensions from several aspects. The topics include Blackwell approachability, PAC-Bayes analysis, probability forecasting with the logarithmic loss, online portfolio selection, and learning with expert advice. |
|
課程目標 |
After taking this course, the students are expected to
- be familiar with basic concepts about Blackwell approachability and the aggregating algorithm and
- be able to read related literature. |
|
課程要求 |
- The students are expected to be motivated enough.
- The students are expected to be familiar with multivariate calculus, probability, and linear algebra. Knowledge of machine learning and statistics may be helpful but are not necessary. |
|
預期每週課後學習時數 |
|
|
Office Hours |
|
|
指定閱讀 |
|
|
參考書目 |
- N. Cesa-Bianchi & G. Lugosi. Prediction, Learning, and Games. 2006.
- J.-F. Mertens et al. Repeated Games. 2015. |
|
評量方式
(僅供參考) |
|
|