Table of Contents

Class AveragedGEM<T>

Namespace
AiDotNet.ContinualLearning
Assembly
AiDotNet.dll

Implements Averaged Gradient Episodic Memory (A-GEM) for continual learning.

public class AveragedGEM<T> : IContinualLearningStrategy<T>

Type Parameters

T

The numeric type for calculations.

Inheritance
AveragedGEM<T>
Implements
Inherited Members

Remarks

For Beginners: A-GEM is a more efficient version of GEM that uses a single random sample from the episodic memory instead of checking constraints against all stored examples. This makes it much faster while maintaining similar performance.

How it works:

  1. Store examples from each completed task in episodic memory.
  2. For each gradient update, compute a reference gradient from a random batch of ALL stored memories (not per-task).
  3. If the proposed gradient would hurt past tasks (negative dot product with reference), project it to be orthogonal.

Key Difference from GEM:

  • GEM: Checks constraints for each task separately (expensive QP solver).
  • A-GEM: Single averaged constraint from random memory sample (simple projection).

Projection Formula:

If g · g_ref < 0: g_proj = g - (g · g_ref / g_ref · g_ref) × g_ref

Reference: Chaudhry, A. et al. "Efficient Lifelong Learning with A-GEM" (2019). ICLR.

Constructors

AveragedGEM(int, int, double, int?)

Initializes a new instance of the AveragedGEM class.

public AveragedGEM(int memorySize = 256, int sampleSize = 64, double lambda = 1, int? seed = null)

Parameters

memorySize int

Maximum samples to store per task (default: 256).

sampleSize int

Batch size to sample from memory for reference gradient (default: 64).

lambda double

Regularization strength for compatibility (default: 1.0).

seed int?

Random seed for reproducibility (default: null for random).

Properties

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.

TaskCount

Gets the number of tasks stored in episodic memory.

public int TaskCount { get; }

Property Value

int

TotalMemorySize

Gets the total number of samples in episodic memory.

public int TotalMemorySize { get; }

Property Value

int

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.

Edit this page