io.github.srs.model.entity.dynamicentity.agent.reward
Members list
Type members
Classlikes
Attributes
- Supertypes
-
class Objecttrait Matchableclass Any
- Self type
-
ExplorationReward.type
Default placeholder reward model. Returns zero for every transition — suitable as a neutral fallback.
Default placeholder reward model. Returns zero for every transition — suitable as a neutral fallback.
Attributes
- Supertypes
-
trait Serializabletrait Producttrait Equalstrait RewardModel[Agent]class Objecttrait Matchableclass AnyShow all
Attributes
- Supertypes
-
class Objecttrait Matchableclass Any
- Self type
Phototaxis Reward Module
Phototaxis Reward Module
Attributes
- Supertypes
-
class Objecttrait Matchableclass Any
- Self type
-
PhototaxisReward.type
An enumeration of available reward models for agents.
An enumeration of available reward models for agents.
This enum represents the different types of reward models that can be applied to agents in the simulation. Each case corresponds to a specific reward calculation strategy.
Attributes
- Supertypes
-
trait Enumtrait Serializabletrait Producttrait Equalsclass Objecttrait Matchableclass AnyShow all
Defines a reward model for evaluating the outcomes of actions performed by entities within an environment.
Defines a reward model for evaluating the outcomes of actions performed by entities within an environment.
Type parameters
- E
-
The type of entity for which the reward model is applicable.
Attributes
- Supertypes
-
class Objecttrait Matchableclass Any
- Known subtypes
-
class NoRewardclass DQObstacleAvoidanceclass QObstacleAvoidanceclass PhototaxisDQclass PhototaxisQclass ExplorationDQNclass ExplorationQLShow all
Manager that maintains reward states for multiple entities.
Manager that maintains reward states for multiple entities.
Attributes
- Supertypes
-
class Objecttrait Matchableclass Any
A trait for defining stateful reward models for entities interacting with an environment.
A trait for defining stateful reward models for entities interacting with an environment.
This trait extends the RewardModel and introduces a state mechanism to maintain contextual information across evaluations, allowing for more complex reward computations based on the dynamic state of the system.
Type parameters
- E
-
The type of entity for which the reward model is applicable.
- S
-
The type of the state maintained by the reward model.
Attributes
- Supertypes
- Known subtypes
-
class ExplorationDQNclass ExplorationQL