Results for "sensitivity to data"

Maintaining alignment under new conditions.

Task instruction without examples.

Applying learned patterns incorrectly.

Centralized AI expertise group.

Devices measuring physical quantities (vision, lidar, force, IMU, etc.).

External sensing of surroundings (vision, audio, lidar).

High-fidelity virtual model of a physical system.

Randomizing simulation parameters to improve real-world transfer.

Differences between simulated and real physics.

RL using learned or known environment models.

Estimating robot position within a map.

AI systems assisting clinicians with diagnosis or treatment decisions.

AI-assisted review of legal documents.

Differences between training and deployed patient populations.

AI predicting crime patterns (highly controversial).

Predicting case success probabilities.

Requirement to reveal AI usage in legal decisions.

AI applied to scientific problems.

AI proposing scientific hypotheses.

Training one model on multiple tasks simultaneously to improve generalization through shared structure.

Configuration choices not learned directly (or not typically learned) that govern training or architecture.

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

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

A gradient method using random minibatches for efficient training on large datasets.

Halting training when validation performance stops improving to reduce overfitting.

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

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

Mechanism that computes context-aware mixtures of representations; scales well and captures long-range dependencies.

Attention where queries/keys/values come from the same sequence, enabling token-to-token interactions.

The text (and possibly other modalities) given to an LLM to condition its output behavior.