Results for "statistical learning"

AI-assisted review of legal documents.

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

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

Framework for reasoning about cause-effect relationships beyond correlation, often using structural assumptions and experiments.

A hidden variable influences both cause and effect, biasing naive estimates of causal impact.

Logging hyperparameters, code versions, data snapshots, and results to reproduce and compare experiments.

Ability to replicate results given same code/data; harder in distributed training and nondeterministic ops.

Converts logits to probabilities by exponentiation and normalization; common in classification and LMs.

Measures how one probability distribution diverges from another.

Bayesian parameter estimation using the mode of the posterior distribution.

Detecting unauthorized model outputs or data leaks.

Maps audio signals to linguistic units.

Shift in feature distribution over time.

Describes likelihoods of random variable outcomes.

Variable whose values depend on chance.

Average value under a distribution.

Measure of spread around the mean.

Measures joint variability between variables.

Normalized covariance.

AI systems assisting clinicians with diagnosis or treatment decisions.

Learning physical parameters from data.

Grouping patients by predicted outcomes.

Predicting disease progression or survival.

Unequal performance across demographic groups.

Differences between training and deployed patient populations.

AI predicting crime patterns (highly controversial).

Models estimating recidivism risk.

Predicting case success probabilities.

Predicting borrower default risk.

Ensuring models comply with lending fairness laws.