TimQu
|
f2035523af
|
Fixed a bug in rewardModel's reduceToStateBasedRewards. Also added a function to check whether all rewards are zero.
Former-commit-id: 4dabd07c66
|
9 years ago |
sjunges
|
57dffbcd4f
|
further operators on models and reward models
Former-commit-id: 0146278302
|
10 years ago |
|
sjunges
|
a28ebac022
|
outcommented code for maximals
Former-commit-id: 096fa39fea
|
10 years ago |
|
sjunges
|
a07c2cacc4
|
missed files
Former-commit-id: 9a7f1ac4a9
|
10 years ago |
|
sjunges
|
cf10fe206a
|
modifiers for standard rew model, fix with non-const overload in model
Former-commit-id: 25348872ee
|
10 years ago |
|
sjunges
|
98162d1d7e
|
interface for rew. model extended for reinforcement learning
Former-commit-id: b69474fc4f
|
10 years ago |
dehnert
|
7946483f09
|
added function to modify state-action rewards in nondeterministic model
Former-commit-id: d3a4c38707
|
10 years ago |
|
dehnert
|
21d9e91586
|
work towards interval reward model
Former-commit-id: 24f7e9684f
|
10 years ago |
|
dehnert
|
b3178e17f6
|
more bug fixes
Former-commit-id: 0b33b30efa
|
10 years ago |
|
dehnert
|
e631dbd1a0
|
more work on new reward models
Former-commit-id: 661c7b5b24
|
10 years ago |
|
dehnert
|
61fb277024
|
more work on refactoring (storm stinks and should be rewritten :P)
Former-commit-id: 7495dea2df
|
10 years ago |
|
dehnert
|
6fa1078fb1
|
some more work on reward model
Former-commit-id: 8533357f34
|
10 years ago |
|
dehnert
|
dcd42d5653
|
started reworking reward models
Former-commit-id: 49d9106451
|
10 years ago |
