Class ExperienceReplay<T>

Namespace

AiDotNet.ContinualLearning

Assembly

AiDotNet.dll

Implements Experience Replay for continual learning.

public class ExperienceReplay<T> : IContinualLearningStrategy<T>

Type Parameters

T

The numeric type for calculations.

Inheritance

object

ExperienceReplay<T>

Implements

IContinualLearningStrategy<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: Experience Replay is one of the simplest and most effective continual learning strategies. It stores examples from previous tasks and mixes them with new task data during training, directly rehearsing old knowledge.

How it works:

1. After each task, store a subset of examples in a replay buffer.
2. During training on new tasks, sample from the buffer and train on both new data and replayed old data.
3. This directly prevents forgetting by repeatedly practicing old tasks.

Buffer Strategies:

• Reservoir: Each sample has equal probability of being kept.
• Ring: FIFO queue, oldest samples are removed first.
• ClassBalanced: Maintains equal samples per class.

Advantages:

• Simple to implement and understand.
• Very effective in practice.
• Can be combined with other strategies.

Reference: Ratcliff, R. "Connectionist models of recognition memory" (1990). Psychological Review. (Original concept); Rolnick et al. "Experience Replay for Continual Learning" (2019). NeurIPS. (Modern application)

Constructors

ExperienceReplay(int, double, BufferStrategy, double, int?)

Initializes a new instance of the ExperienceReplay class.

public ExperienceReplay(int maxBufferSize = 1000, double replayRatio = 0.5, ExperienceReplay<T>.BufferStrategy strategy = BufferStrategy.Reservoir, double lambda = 1, int? seed = null)

Parameters

maxBufferSize int

Maximum samples to store in buffer (default: 1000).

replayRatio double

Ratio of replay samples to new samples (default: 0.5).

strategy ExperienceReplay<T>.BufferStrategy

Buffer management strategy (default: Reservoir).

lambda double

Weight for replay loss contribution (default: 1.0).

seed int?

Random seed for reproducibility (default: null).

Properties

BufferSize

Gets the current buffer size.

public int BufferSize { get; }

Property Value

int

Lambda

Gets the regularization strength parameter (lambda) for loss-based continual learning.

public double Lambda { get; set; }

Property Value

double

Remarks

For Beginners: Lambda controls how strongly the strategy prevents forgetting. A higher lambda means the network is more conservative about changing weights important for previous tasks, but this might make it harder to learn new tasks effectively.

Typical values range from 100 to 10000, depending on the complexity of tasks and how important it is to preserve old knowledge versus learning new knowledge.

ReplayRatio

Gets the replay ratio.

public double ReplayRatio { get; }

Property Value

double

Strategy

Gets the buffer strategy.

public ExperienceReplay<T>.BufferStrategy Strategy { get; }

Property Value

ExperienceReplay<T>.BufferStrategy

Methods

AfterTask(INeuralNetwork<T>, (Tensor<T> inputs, Tensor<T> targets), int)

Processes information after completing training on a task.

public void AfterTask(INeuralNetwork<T> network, (Tensor<T> inputs, Tensor<T> targets) taskData, int taskId)

Parameters

network INeuralNetwork<T>

The neural network that was trained.

taskData (Tensor<T> grad1, Tensor<T> grad2)

Data from the completed task for computing importance measures.

taskId int

The identifier for the completed task.

Remarks

For Beginners: This method is called after you finish training on a task. It allows the strategy to compute and store information about what the network learned, which will be used to protect this knowledge when learning future tasks.

For example, in Elastic Weight Consolidation (EWC), this computes the Fisher Information Matrix to identify which weights are most important for the completed task.

BeforeTask(INeuralNetwork<T>, int)

Prepares the strategy before starting to learn a new task.

public void BeforeTask(INeuralNetwork<T> network, int taskId)

Parameters

network INeuralNetwork<T>

The neural network that will be trained.

taskId int

The identifier for the upcoming task (0-indexed).

Remarks

For Beginners: This method is called before you start training on a new task. It allows the strategy to capture the network's current state or prepare any necessary data structures for protecting knowledge from previous tasks.

For example, in Learning without Forgetting (LwF), this might store the network's predictions on the new task's inputs before training begins, so we can later encourage the network to maintain similar predictions.

ComputeLoss(INeuralNetwork<T>)

Computes the regularization loss to prevent forgetting previous tasks.

public T ComputeLoss(INeuralNetwork<T> network)

Parameters

network INeuralNetwork<T>

The neural network being trained.

Returns

T

The regularization loss value that should be added to the task loss.

Remarks

For Beginners: This method calculates an additional loss term that penalizes the network for deviating from its learned knowledge of previous tasks. You add this to your regular task loss during training:

var totalLoss = taskLoss + strategy.ComputeLoss(network);

For example, in EWC, this returns a penalty proportional to how much important weights have changed from their optimal values for previous tasks. Larger changes to important weights result in higher loss, discouraging the network from forgetting.

ModifyGradients(INeuralNetwork<T>, Vector<T>)

Modifies the gradient to prevent catastrophic forgetting.

public Vector<T> ModifyGradients(INeuralNetwork<T> network, Vector<T> gradients)

Parameters

network INeuralNetwork<T>

The neural network being trained.

gradients Vector<T>

The gradients from the current task loss.

Returns

Vector<T>

Modified gradients that protect previous task knowledge.

Remarks

For Beginners: Some continual learning strategies work by modifying the gradients (the update directions for weights) rather than adding a loss term. This method takes the gradients computed from the current task and modifies them to avoid interfering with previously learned tasks.

For example, in Gradient Episodic Memory (GEM), if a gradient would hurt performance on stored examples from previous tasks, it's projected to the closest gradient that doesn't interfere with those examples.

If a strategy doesn't use gradient modification, this should return the gradients unchanged.

Reset()

Resets the strategy, clearing all stored task information.

public void Reset()

Remarks

For Beginners: This method clears all the information the strategy has accumulated about previous tasks. After calling this, the network will be free to learn new tasks without any constraints from previously learned tasks.

Use this when you want to start fresh or when you're done with a sequence of tasks and want to begin a new independent sequence.

SampleMixedBatch(Tensor<T>, Tensor<T>, int)

Samples a mixed batch combining current task data with replay data.

public (Tensor<T> inputs, Tensor<T> targets) SampleMixedBatch(Tensor<T> currentInputs, Tensor<T> currentTargets, int batchSize)

Parameters

currentInputs Tensor<T>

currentTargets Tensor<T>

batchSize int

Returns

(Tensor<T> grad1, Tensor<T> grad2)

SampleReplayBatch()

Samples a batch from the replay buffer for training.

public (Tensor<T> inputs, Tensor<T> targets) SampleReplayBatch()

Returns

(Tensor<T> grad1, Tensor<T> grad2)

Tuple of inputs and targets tensors.