Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Parsing for POMDP with state rewards fails #689

Open
TheGreatfpmK opened this issue Mar 17, 2025 · 0 comments
Open

Parsing for POMDP with state rewards fails #689

TheGreatfpmK opened this issue Mar 17, 2025 · 0 comments

Comments

@TheGreatfpmK
Copy link
Contributor

When parsing a POMDP with state rewards, the state rewards are ignored i.e. no reward model for state rewards is created, and the rewards (at least when parsing DRN) are parsed as state labels instead. For example, on this POMDP pomdp-state-rewards.zip I get following output after parsing:

Model type:     POMDP (sparse)
States:         16
Transitions:    26
Choices:        18
Observations:   5
Reward Models:  none
State Labels:   5 labels
   * deadlock -> 2 item(s)
   * [100] -> 2 item(s)
   * init -> 1 item(s)
   * done -> 2 item(s)
   * [0] -> 14 item(s)
Choice Labels:  5 labels
   * end -> 4 item(s)
   * eval -> 4 item(s)
   * roll2 -> 4 item(s)
   * roll1better -> 2 item(s)
   * roll1 -> 2 item(s)

as you can see the state rewards [100] and [0] were converted to state labels. It seems like there's a bug in the parser for POMDPs at least. I tried it with MDPs and there the state rewards are parsed correctly.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant