Class RewardClippingConfig<T>

Namespace

AiDotNet.Configuration

Assembly

AiDotNet.dll

Configuration for reward clipping.

public class RewardClippingConfig<T>

Type Parameters

T

The numeric type.

Inheritance

object

RewardClippingConfig<T>

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

For Beginners: Clipping rewards to a fixed range can stabilize training when reward magnitudes vary widely. The famous Atari DQN paper clipped rewards to [-1, 1].

Constructors

RewardClippingConfig()

Creates a new instance with default values.

public RewardClippingConfig()

Properties

MaxReward

Maximum reward value after clipping.

public T MaxReward { get; set; }

Property Value

T

MinReward

Minimum reward value after clipping.

public T MinReward { get; set; }

Property Value

T

UseClipping

Whether to clip rewards (vs just scaling them).

public bool UseClipping { get; set; }

Property Value

bool