Table of Contents

Class IQLAgent<T>

Namespace
AiDotNet.ReinforcementLearning.Agents.IQL
Assembly
AiDotNet.dll

Implicit Q-Learning (IQL) agent for offline reinforcement learning.

public class IQLAgent<T> : DeepReinforcementLearningAgentBase<T>, IRLAgent<T>, IFullModel<T, Vector<T>, Vector<T>>, IModel<Vector<T>, Vector<T>, ModelMetadata<T>>, IModelSerializer, ICheckpointableModel, IParameterizable<T, Vector<T>, Vector<T>>, IFeatureAware, IFeatureImportance<T>, ICloneable<IFullModel<T, Vector<T>, Vector<T>>>, IGradientComputable<T, Vector<T>, Vector<T>>, IJitCompilable<T>, IDisposable

Type Parameters

T

The numeric type used for calculations.

Inheritance
IQLAgent<T>
Implements
IFullModel<T, Vector<T>, Vector<T>>
IModel<Vector<T>, Vector<T>, ModelMetadata<T>>
IParameterizable<T, Vector<T>, Vector<T>>
ICloneable<IFullModel<T, Vector<T>, Vector<T>>>
IGradientComputable<T, Vector<T>, Vector<T>>
Inherited Members
Extension Methods

Remarks

IQL uses expectile regression to learn a value function that focuses on high-return trajectories, enabling effective offline policy learning without explicit conservative penalties like CQL.

For Beginners: IQL is an offline RL algorithm that learns from fixed datasets. It uses a clever statistical technique (expectile regression) to avoid overestimating values of unseen actions.

Key features:

  • Expectile Regression: Asymmetric loss that focuses on upper quantiles
  • Three Networks: V(s), Q(s,a), and π(a|s)
  • Simpler than CQL: No conservative penalties or Lagrangian multipliers
  • Advantage-Weighted Regression: Extracts policy from Q and V functions

Think of expectiles like percentiles - focusing on "typically good" outcomes rather than "best possible" outcomes helps avoid overoptimism.

Advantages:

  • Simpler hyperparameter tuning than CQL
  • Often more stable
  • Good for offline datasets with diverse quality

Constructors

IQLAgent(IQLOptions<T>) 

public IQLAgent(IQLOptions<T> options)

Parameters

options IQLOptions<T>

Properties

FeatureCount

Gets the number of input features (state dimensions).

public override int FeatureCount { get; }

Property Value

int

Methods

ApplyGradients(Vector<T>, T)

Applies gradients to update the agent.

public override void ApplyGradients(Vector<T> gradients, T learningRate)

Parameters

gradients Vector<T>
learningRate T

Clone()

Clones the agent.

public override IFullModel<T, Vector<T>, Vector<T>> Clone()

Returns

IFullModel<T, Vector<T>, Vector<T>>

ComputeGradients(Vector<T>, Vector<T>, ILossFunction<T>?)

Computes gradients for the agent.

public override Vector<T> ComputeGradients(Vector<T> input, Vector<T> target, ILossFunction<T>? lossFunction = null)

Parameters

input Vector<T>
target Vector<T>
lossFunction ILossFunction<T>

Returns

Vector<T>

Deserialize(byte[])

Deserializes the agent from bytes.

public override void Deserialize(byte[] data)

Parameters

data byte[]

GetMetrics()

Gets the current training metrics.

public override Dictionary<string, T> GetMetrics()

Returns

Dictionary<string, T>

Dictionary of metric names to values.

GetModelMetadata()

Gets model metadata.

public override ModelMetadata<T> GetModelMetadata()

Returns

ModelMetadata<T>

GetParameters()

Gets the agent's parameters.

public override Vector<T> GetParameters()

Returns

Vector<T>

LoadModel(string)

Loads the agent's state from a file.

public override void LoadModel(string filepath)

Parameters

filepath string

Path to load the agent from.

LoadOfflineData(List<(Vector<T> state, Vector<T> action, T reward, Vector<T> nextState, bool done)>)

Load offline dataset into the replay buffer.

public void LoadOfflineData(List<(Vector<T> state, Vector<T> action, T reward, Vector<T> nextState, bool done)> dataset)

Parameters

dataset List<(Vector<T> state, Vector<T> action, T reward, Vector<T> nextState, bool done)>

Predict(Vector<T>)

Makes a prediction using the trained agent.

public override Vector<T> Predict(Vector<T> input)

Parameters

input Vector<T>

Returns

Vector<T>

PredictAsync(Vector<T>) 

public Task<Vector<T>> PredictAsync(Vector<T> input)

Parameters

input Vector<T>

Returns

Task<Vector<T>>

ResetEpisode()

Resets episode-specific state (if any).

public override void ResetEpisode()

SaveModel(string)

Saves the agent's state to a file.

public override void SaveModel(string filepath)

Parameters

filepath string

Path to save the agent.

SelectAction(Vector<T>, bool)

Selects an action given the current state observation.

public override Vector<T> SelectAction(Vector<T> state, bool training = true)

Parameters

state Vector<T>

The current state observation as a Vector.

training bool

Whether the agent is in training mode (affects exploration).

Returns

Vector<T>

Action as a Vector (can be discrete or continuous).

Serialize()

Serializes the agent to bytes.

public override byte[] Serialize()

Returns

byte[]

SetParameters(Vector<T>)

Sets the agent's parameters.

public override void SetParameters(Vector<T> parameters)

Parameters

parameters Vector<T>

StoreExperience(Vector<T>, Vector<T>, T, Vector<T>, bool)

Stores an experience tuple for later learning.

public override void StoreExperience(Vector<T> state, Vector<T> action, T reward, Vector<T> nextState, bool done)

Parameters

state Vector<T>

The state before action.

action Vector<T>

The action taken.

reward T

The reward received.

nextState Vector<T>

The state after action.

done bool

Whether the episode terminated.

Train()

Performs one training step, updating the agent's policy/value function.

public override T Train()

Returns

T

The training loss for monitoring.

TrainAsync() 

public Task TrainAsync()

Returns

Task
Edit this page