Results for "learning rate"

Adjusting learning rate over training to improve convergence.

Controls the size of parameter updates; too high diverges, too low trains slowly or gets stuck.

Gradually increasing learning rate at training start to avoid divergence.

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

Rate at which AI capabilities improve.

Iterative method that updates parameters in the direction of negative gradient to minimize loss.

Methods like Adam adjusting learning rates dynamically.

Ability to correctly detect disease.

Failure to detect present disease.

Organizational uptake of AI technologies.

Maximum system processing rate.

Trend reversal when data is aggregated improperly.

A model is PAC-learnable if it can, with high probability, learn an approximately correct hypothesis from finite samples.

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

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

Model relies on irrelevant signals.

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

How many requests or tokens can be processed per unit time; affects scalability and cost.

Converting audio speech into text, often using encoder-decoder or transducer architectures.

Direction of steepest ascent of a function.

Of true positives, the fraction correctly identified; sensitive to false negatives.

Of true negatives, the fraction correctly identified.

Classical controller balancing responsiveness and stability.

Mathematical representation of friction forces.

Returns above benchmark.

Ordering training samples from easier to harder to improve convergence or generalization.

Selecting the most informative samples to label (e.g., uncertainty sampling) to reduce labeling cost.

Learning from data generated by a different policy.

Learning a function from input-output pairs (labeled data), optimizing performance on predicting outputs for unseen inputs.

Learning policies from expert demonstrations.