ExpectedSARSAAgent<T>

Models & Types · AiDotNet.ReinforcementLearning.Agents.ExpectedSARSA

Expected SARSA agent using tabular methods.

Inheritance: ReinforcementLearningAgentBase<T> → ExpectedSARSAAgent<T>

Implements: IGradientComputable<T, Vector<T>, Vector<T>>

For Beginners

Expected SARSA is like SARSA but instead of using the actual next action, it uses the average Q-value weighted by the probability of taking each action. This reduces variance compared to SARSA.

Update: Q(s,a) ← Q(s,a) + α[r + γ Σ π(a’|s’)Q(s’,a’) - Q(s,a)]

Benefits over SARSA:

  • Lower Variance: Averages over actions instead of sampling
  • Off-Policy Learning: Can learn optimal policy while exploring
  • Better Performance: Often converges faster than SARSA

Famous for: Van Seijen et al. 2009, bridging SARSA and Q-Learning

How It Works

Expected SARSA is a TD control algorithm that uses the expected value under the current policy instead of sampling the next action.

Example

// Create an Expected SARSA agent that averages over next actions
var options = new ExpectedSARSAOptions<double>(stateSize: 4, actionSize: 2) { LearningRate = 0.1 };
var agent = new ExpectedSARSAAgent<double>(options);
            
// Select an action for the current state
var state = new Vector<double>(new double[] { 0.5, -0.3, 1.0, 0.2 });
var action = agent.SelectAction(state);

Constructors

Constructor Summary
ExpectedSARSAAgent() Initializes a new instance with default settings.
ExpectedSARSAAgent(ExpectedSARSAOptions<T> options)

Properties

Property Summary
int FeatureCount { get; }

Methods

Method Summary
ModelMetadata<T> GetModelMetadata()
T Train()
Vector<T> ComputeGradients(Vector<T> input, Vector<T> target, ILossFunction<T> lossFunction)
Vector<T> SelectAction(Vector<T> state, bool training)
byte[] Serialize() Auto-generated common model serialization surface.
void ApplyGradients(Vector<T> gradients, T learningRate)
void Deserialize(byte[] data) Auto-generated common model deserialization surface.
void LoadModel(string filepath)
void SaveModel(string filepath)
void StoreExperience(Vector<T> state, Vector<T> action, T reward, Vector<T> nextState, bool done)