Results for "shortcut learning"

AI selecting next experiments.

AI tacitly coordinating prices.

Rate at which AI capabilities improve.

Research ensuring AI remains safe.

A conceptual framework describing error as the sum of systematic error (bias) and sensitivity to data (variance).

Updating a pretrained model’s weights on task-specific data to improve performance or adapt style/behavior.

Measures a model’s ability to fit random noise; used to bound generalization error.

A continuous vector encoding of an item (word, image, user) such that semantic similarity corresponds to geometric closeness.

The learned numeric values of a model adjusted during training to minimize a loss function.

A scalar measure optimized during training, typically expected loss over data, sometimes with regularization terms.

Minimizing average loss on training data; can overfit when data is limited or biased.

When a model fits noise/idiosyncrasies of training data and performs poorly on unseen data.

When a model cannot capture underlying structure, performing poorly on both training and test data.

How well a model performs on new data drawn from the same (or similar) distribution as training.

Separating data into training (fit), validation (tune), and test (final estimate) to avoid leakage and optimism bias.

A robust evaluation technique that trains/evaluates across multiple splits to estimate performance variability.

A table summarizing classification outcomes, foundational for metrics like precision, recall, specificity.

Plots true positive rate vs false positive rate across thresholds; summarizes separability.

Scalar summary of ROC; measures ranking ability, not calibration.

A proper scoring rule measuring squared error of predicted probabilities for binary outcomes.

Penalizes confident wrong predictions heavily; standard for classification and language modeling.

Average of squared residuals; common regression objective.

Uses an exponential moving average of gradients to speed convergence and reduce oscillation.

One complete traversal of the training dataset during training.

Halting training when validation performance stops improving to reduce overfitting.

A parameterized function composed of interconnected units organized in layers with nonlinear activations.

Methods to set starting weights to preserve signal/gradient scales across layers.

Nonlinear functions enabling networks to approximate complex mappings; ReLU variants dominate modern DL.

Randomly zeroing activations during training to reduce co-adaptation and overfitting.

Networks using convolution operations with weight sharing and locality, effective for images and signals.