---------------
-- ToDo list --
---------------

P0
- Create framework for HRL
	- MAXQ algorithm						... to be tested
	- 2-rooms and/or 4-rooms tasks
		- Article from Botvinick et al on Drive: look at their implementation of 4-rooms
		- Implement 				2-rooms OK, Q-learning & value iteration work
- Nice presentation of results in Notebook 	Q-learning + MAXQ on 2-room is done

P1
- Add option discovery to the framework		... methods done, disconnected from MAXQ
	- See OGovern, chapter 4 for the theory and applications to the 2-room task 	OK
- Presentation in Notebook 	...