Difference between revisions of "CS4646 Spring 2018"
Jump to navigation
Jump to search
Line 11: | Line 11: | ||
* Forum: https://piazza.com/class/jby8m47v3v52oc (just search for CS 4646 if this doesn't work) | * Forum: https://piazza.com/class/jby8m47v3v52oc (just search for CS 4646 if this doesn't work) | ||
− | * Schedule: | + | * Schedule: Subject to Change if necessary. I will give you as much notice as possible. |
Revision as of 14:44, 10 January 2018
DRAFT
This page is still in progress. Don't consider it final until the first day of class.
Overview
You are on the page for information specific to the Spring 2018 session of CS 4646. Go here (Undergrad_ML4T) for overall course policies.
Schedule & Forum
- Forum: https://piazza.com/class/jby8m47v3v52oc (just search for CS 4646 if this doesn't work)
- Schedule: Subject to Change if necessary. I will give you as much notice as possible.
Week | Date (Tues) | Weekly Topics | Due |
---|---|---|---|
1 | 1/9 | Course Overview, Python/Pandas Tutorial | |
2 | 1/16 | Numpy Tutorial, Visualizing Market Data, Working with Time Series, Incomplete Data | |
3 | 1/23 | ML Lexicon/Taxonomy, Evaluating Learners | Project 1 |
4 | 1/30 | Supervised Learning (KNN, LinReg, Decision Trees) | |
5 | 2/6 | Ensembles (Bagging, Boosting) | |
6 | 2/13 | Market History, Actors, Order Types | Project 2 |
7 | 2/20 | Order Book, Leverage, Valuation | |
8 | 2/27 | Technical Analysis, Candlestick Chart Patterns, Time Value of Money, CAPM | Project 3 |
9 | 3/6 | Efficient Market Hypothesis, Fund. Law of APM, Efficient Frontier | |
10 | 3/13 | Review, Exam 1 | Exam 1 |
11 | 3/20 | Finite Automata, MDP, Value/Policy Iteration | Project 4 |
12 | 3/27 | Q-Learning | |
13 | 4/3 | Misc Machine Learning / Catch-up | Project 5 |
14 | 4/10 | Options, Time Series Q-Learning | |
15 | 4/17 | Review, Exam 2, Last regular day of class | Exam 2 |
16 | 4/24 | Final Instruction Days (Tuesday), Help Session on Final Project | Project 6 |
17 | 5/1 | Finals week, NO FINAL EXAM |
Assignments
Projects (60%)
- [assess_portfolio] 5% (easy)
- Regression / Ensemble Learners 10% (challenging)
- Market Simulator 10% (moderate)
- Manual Strategy 10% (moderate)
- Q-Learning Robot 10% (moderate)
- Strategy Learner 15% (very challenging)
Exams (40%)
- Exam 1: 20%
- Exam 2: 20%
Exam Study Guides
Thresholds
- A: 90% and above
- B: 80% and above
- C: 70% and above
- D: 60% and above
- F: below 60%
These are hard boundaries (we round down).