Results for "learning signal"

Embedding signals to prove model ownership.

Detects trigger phrases in audio streams.

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

Modifying reward to accelerate learning.

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

Optimization under uncertainty.

Using production outcomes to improve models.

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

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

Maps audio signals to linguistic units.

Identifying speakers in audio.

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

Software pipeline converting raw sensor data into structured representations.

Learning physical parameters from data.

Aligns transcripts with audio timestamps.

Models time evolution via hidden states.

Monte Carlo method for state estimation.

Maximizing reward without fulfilling real goal.

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

Control without feedback after execution begins.

Signals indicating dangerous behavior.

Using markers to isolate context segments.

Adjusting learning rate over training to improve convergence.

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.

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

Learning policies from expert demonstrations.

Learning where data arrives sequentially and the model updates continuously, often under changing distributions.

Learning structure from unlabeled data, such as discovering groups, compressing representations, or modeling data distributions.