CS7646 Summer 2017
You are on the page for information specific to the Summer 2017 session of this course. Go here (Machine_Learning_for_Trading_Course) for overall course policies.
Specific On-Campus Information
- Any on-campus deviations (related to office hours, exams, lecture times/topics/notes) can be found here: CS7646_Summer_2017_ATL
2017 Summer Schedule (OMS)
This is a link to the schedule for the online (OMS) session of the course. The project deadlines are the same for the on campus session.
Assignments & Grading
- [MC1-Project-1: Assess portfolio] 5% (easy)
- [MC3-Project-1: Implement and assess a regression learner using decision trees and random forests] 15% (hard)
- [MC2-Project-1: Build a market simulator] 15% (moderate)
- [MC3-Project-2: Q-learning maze navigation] 10% (easy)
- Trading Strategy Learner 17% (very hard)
- For this project you will implement a machine learning based trading system. You can choose whether you want to use a decision-tree based approach or a Q-learning based approach. This project is being reformulated, but you can follow the links below to previous versions of these two projects to get a feel for what the requirements will be:
- [MC3-Project-3: Implement a "manual" quant strategy, then do it with decision tree classification]
- [MC3-Project-4: Q-learning trader]
Class participation (3%)
- Class participation is determined by activity on piazza.
- 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).