Skip to content

Ondra fix reward assignment #481

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.

Sign up for GitHub

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

Merged
ondrej-lukas merged 2 commits into from
Mar 5, 2026
Merged

Ondra fix reward assignment #481

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e5fc75e
Add check for negative rewards
ondrej-lukas Mar 5, 2026
010526d
Fix the sign of FP reward (assume non-positive value)
ondrej-lukas Mar 5, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 8 additions & 1 deletion netsecgame/game/configuration_manager.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -134,7 +134,14 @@ def get_rewards(self, reward_names: List[str] = ["step", "success", "fail", "fal
"""Returns the rewards configuration."""
if not self._parser:
raise RuntimeError("Configuration not loaded.")
return self._parser.get_rewards(reward_names, default_value)
rewards = self._parser.get_rewards(reward_names, default_value)
if rewards.get("fail") > 0:
self.logger.warning("Fail reward is positive. This is not recommended.")
if rewards.get("false_positive") > 0:
self.logger.warning("False positive reward is positive. This is not recommended.")
if rewards.get("success") < 0:
self.logger.warning("Success reward is negative. This is not recommended.")
return rewards

def get_use_dynamic_ips(self, default_value: bool = False) -> bool:
if not self._parser:
Expand Down
2 changes: 1 addition & 1 deletion netsecgame/game/coordinator.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -560,7 +560,7 @@ async def _assign_rewards_episode_end(self):
self._agent_status[agent] = AgentStatus.Fail
# dicrease the reward for false positives
self.logger.debug(f"Processing false positives for agent {agent}: {self._agent_false_positives[agent]}")
self._agent_rewards[agent] -= self._agent_false_positives[agent] * self._rewards["false_positive"]
self._agent_rewards[agent] += self._agent_false_positives[agent] * self._rewards["false_positive"]
# clear the episode end event
self._episode_end_event.clear()
# notify all waiting agents
Expand Down