Tree [r4] /
History



File Date Author Commit
doc 2007-11-12 berger [r1] initial import 1.2 alpha
src 2007-11-12 berger [r1] initial import 1.2 alpha
AUTHORS 2007-11-12 berger [r1] initial import 1.2 alpha
COPYING 2007-11-12 berger [r1] initial import 1.2 alpha
ChangeLog 2007-11-12 berger [r1] initial import 1.2 alpha
INSTALL 2007-11-12 berger [r1] initial import 1.2 alpha
Makefile.am 2007-11-12 berger [r1] initial import 1.2 alpha
NEWS 2007-11-13 ralfberger [r2]
README 2007-11-17 ralfberger [r4] Anjuta-1 project file added
RL++.prj 2007-11-17 ralfberger [r4] Anjuta-1 project file added
TODO 2007-11-12 berger [r1] initial import 1.2 alpha
autogen.sh 2007-11-12 berger [r1] initial import 1.2 alpha
config.guess 2007-11-12 berger [r1] initial import 1.2 alpha
config.h.in 2007-11-12 berger [r1] initial import 1.2 alpha
config.sub 2007-11-12 berger [r1] initial import 1.2 alpha
configure.in 2007-11-12 berger [r1] initial import 1.2 alpha
copying 2007-11-12 berger [r1] initial import 1.2 alpha
depcomp 2007-11-12 berger [r1] initial import 1.2 alpha
install-sh 2007-11-12 berger [r1] initial import 1.2 alpha
ltmain.sh 2007-11-16 ralfberger [r3]
missing 2007-11-12 berger [r1] initial import 1.2 alpha
mkinstalldirs 2007-11-12 berger [r1] initial import 1.2 alpha
setup-gettext 2007-11-12 berger [r1] initial import 1.2 alpha
stamp-h.in 2007-11-12 berger [r1] initial import 1.2 alpha

Read Me

License: GNU LESSER GENERAL PUBLIC LICENSE --> see COPYING for details!

See INSTALL for some basic information on how to install the library.

I apologize for the missing documentation, but I hope the examples will answer most of your possible questions.
In case of problems feel free to contact me at michaelgollin@users.sourceforge.net
and use the bugtracker and the forum on sf.net.

The package includes:
- all library source files for
  - reinforcement learning algorithms (Sarsa, Watkins Q-Learning, TD --> all with Eligibility Traces)
  - value functions (TileCoding (state and action based), LookupTable (state and action based) and Neuronal Network (only state based) )
  - utility classes
- 5 examples:
  - MountainCar with Actions (TileCoding), States (TileCoding and Neuronal Network)
  - Windy GridWorld Scenario with Actions and States (both Lookup Table)

Written and tested on:
  - Suse >= 8.2, Gentoo >= 2006
  - Kernel >= 2.4.20
  - gcc >= 3.3.1