JoeyAndres rl issues - Githubissues

JoeyAndres / rl

Reinforcement Learning library

2 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Convert underlying class/methods to structs/functions or C.

#40 JoeyAndres opened 7 years ago
0
Softmax suboptimal action choice.

#39 JoeyAndres opened 7 years ago
0
TD-lambda

#38 JoeyAndres opened 7 years ago
0
Config file

#37 JoeyAndres opened 7 years ago
0
Logging

#36 JoeyAndres opened 7 years ago
0
Actuator for known environment.

#35 JoeyAndres opened 7 years ago
1
Unsupervised learning bug in gradient descent.

#34 JoeyAndres closed 7 years ago
0
Gradient Descent step size is divided by number of tilings.

#33 JoeyAndres closed 7 years ago
0
Integrate SDM

#32 JoeyAndres closed 7 years ago
1
Add more discrete Toy-Problem tests.

#31 JoeyAndres opened 7 years ago
1
Problem formulation interface

#30 JoeyAndres opened 7 years ago
0
Make StateActionPairContainer follow the proper way to override stl containers.

#29 JoeyAndres opened 7 years ago
0
Need SarsaGD and QLearningGD

#28 JoeyAndres closed 8 years ago
0
Read /cpu/cpuinfo to autoatically have the appropriate flag.

#27 JoeyAndres opened 8 years ago
0
Make globals in declares.h overridable

#26 JoeyAndres opened 8 years ago
0
GPU computing for updating feature vectors

#25 JoeyAndres opened 8 years ago
0
TileCode: Support for non-grid tiling.

#24 JoeyAndres opened 8 years ago
0
Tile Coding: RBF

#23 JoeyAndres opened 8 years ago
0
Serialization/Deserialization

#22 JoeyAndres closed 7 years ago
1
Null Action for terminal state.

#21 JoeyAndres closed 7 years ago
2
Specialized for integral states and/or action.

#20 JoeyAndres opened 8 years ago
0
Use std:shared_ptr instead of reference.

#19 JoeyAndres closed 8 years ago
0
Run unit tests for other os to know when our dev break in what os.

#18 JoeyAndres closed 8 years ago
0
Pretested Integration

#17 JoeyAndres opened 8 years ago
0
Fix cpp lint issues.

#16 JoeyAndres closed 7 years ago
0
cpplint.py to uphold google c++ style

#15 JoeyAndres closed 7 years ago
0
Introduce folders for submodules to make searching easier.

#14 JoeyAndres closed 8 years ago
0
Remove AI namespace.

#13 JoeyAndres closed 8 years ago
1
Templatize TileCode

#12 JoeyAndres closed 7 years ago
0
Distributed computing.

#11 JoeyAndres opened 8 years ago
0
Gradient Descent modules failing in mac.

#10 JoeyAndres closed 8 years ago
1
Node bindings.

#9 JoeyAndres opened 8 years ago
0
Mongo backend with sharding.

#8 JoeyAndres closed 7 years ago
1
Mongo backend. Non sharding.

#7 JoeyAndres closed 7 years ago
4
Change tests from cppunit to https://github.com/philsquared/Catch

#6 JoeyAndres closed 8 years ago
0
Setup a ci server.

#5 JoeyAndres closed 8 years ago
0
Investigate backend storage.

#4 JoeyAndres closed 8 years ago
1
Sparse Tile Coding.

#3 JoeyAndres opened 8 years ago
1
Handle prediction problems instead of just control problems.

#2 JoeyAndres closed 8 years ago
0
Multi-threading for learning algorithms.

#1 JoeyAndres opened 8 years ago
0