Class SARSAOptions<T>

Namespace

AiDotNet.Models.Options

Assembly

AiDotNet.dll

Configuration options for SARSA agents.

public class SARSAOptions<T> : ReinforcementLearningOptions<T>

Type Parameters

T

The numeric type used for calculations.

Inheritance

object

ReinforcementLearningOptions<T>

SARSAOptions<T>

Inherited Members

ReinforcementLearningOptions<T>.LearningRate

ReinforcementLearningOptions<T>.DiscountFactor

ReinforcementLearningOptions<T>.LossFunction

ReinforcementLearningOptions<T>.Seed

ReinforcementLearningOptions<T>.BatchSize

ReinforcementLearningOptions<T>.ReplayBufferSize

ReinforcementLearningOptions<T>.TargetUpdateFrequency

ReinforcementLearningOptions<T>.UsePrioritizedReplay

ReinforcementLearningOptions<T>.EpsilonStart

ReinforcementLearningOptions<T>.EpsilonEnd

ReinforcementLearningOptions<T>.EpsilonDecay

ReinforcementLearningOptions<T>.WarmupSteps

ReinforcementLearningOptions<T>.MaxGradientNorm

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

SARSA (State-Action-Reward-State-Action) is an on-policy TD control algorithm. Unlike Q-Learning, it updates based on the action actually taken.

For Beginners: SARSA is more conservative than Q-Learning because it learns from actions it actually takes (including exploratory ones). This makes it safer in environments where bad actions can be catastrophic.

Classic example: Cliff Walking

• Q-Learning learns the shortest path (risky, close to cliff)
• SARSA learns a safer path (further from cliff)

Use SARSA when:

• Safety matters during training
• You want to learn a safe policy
• Environment has dangerous states

Use Q-Learning when:

• You want the optimal policy
• Safety during training doesn't matter
• You can afford exploratory mistakes

Properties

ActionSize

Size of the action space (number of possible actions).

public int ActionSize { get; init; }

Property Value

int

StateSize

Size of the state space (number of state features).

public int StateSize { get; init; }

Property Value

int