Free Printable Flashcard Generator

An efficient algorithm for computing the gradient of the loss function with respect to each weight via the mathematical chain rule.

A first-order iterative optimization algorithm for finding a local minimum of a differentiable objective function by taking steps proportional to the negative gradient.

A deep learning architecture that applies learnable spatial filters (kernels) with weight sharing to process grid-structured data such as images.

A sequence modeling mechanism that computes pairwise relevance weights across all tokens simultaneously without recurrence.

Techniques introduced into the objective function to discourage model complexity and prevent overfitting on training data.

A condition where a statistical model fits noise and idiosyncrasies of the training data so closely that its generalization to unseen test data degrades.

A mathematical function that maps an event or values of one or more variables onto a real number intuitively representing the penalty associated with the event.

A mathematical gate added to artificial neurons (e.g., ReLU, Sigmoid, GELU) to introduce non-linearity into network representations.

A performance measurement for classification problems at various threshold settings, plotting True Positive Rate against False Positive Rate.

The process of optimizing configurations external to the model (such as learning rate, batch size, or layer depth) that are not learned directly from training data.