Results for "feature attribution"

Feature attribution method grounded in cooperative game theory for explaining predictions in tabular settings.

Centralized repository for curated features.

Assigning AI costs to business units.

Studying internal mechanisms or input influence on outputs (e.g., saliency maps, SHAP, attention analysis).

A measurable property or attribute used as model input (raw or engineered), such as age, pixel intensity, or token ID.

Designing input features to expose useful structure (e.g., ratios, lags, aggregations), often crucial outside deep learning.

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

A mismatch between training and deployment data distributions that can degrade model performance.

A structured collection of examples used to train/evaluate models; quality, bias, and coverage often dominate outcomes.

Techniques to understand model decisions (global or local), important in high-stakes and regulated settings.

Automatically learning useful internal features (latent variables) that capture salient structure for downstream tasks.

Human or automated process of assigning targets; quality, consistency, and guidelines matter heavily.

When information from evaluation data improperly influences training, inflating reported performance.

A measure of randomness or uncertainty in a probability distribution.

Quantifies shared information between random variables.

Allows gradients to bypass layers, enabling very deep networks.

A narrow hidden layer forcing compact representations.

Simplified Boltzmann Machine with bipartite structure.

Probabilistic graphical model for structured prediction.

Model that compresses input into latent space and reconstructs it.

Combining signals from multiple modalities.

Changing speaker characteristics while preserving content.

Identifying speakers in audio.

CNNs applied to time series.

End-to-end process for model training.

Shift in feature distribution over time.

Running predictions on large datasets periodically.

Models whose weights are publicly available.

Vector whose direction remains unchanged under linear transformation.

Normalized covariance.