Results for "statistical learning"

Methods that learn training procedures or initializations so models can adapt quickly to new tasks with little data.

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

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

Learning only from current policy’s data.

A theoretical framework analyzing what classes of functions can be learned, how efficiently, and with what guarantees.

Learning from data by constructing “pseudo-labels” (e.g., next-token prediction, masked modeling) without manual annotation.

Achieving task performance by providing a small number of examples inside the prompt without weight updates.

Training across many devices/silos without centralizing raw data; aggregates updates, not data.

Gradually increasing learning rate at training start to avoid divergence.

Built-in assumptions guiding learning efficiency and generalization.

Modifying reward to accelerate learning.

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

Reinforcement learning from human feedback: uses preference data to train a reward model and optimize the policy.

Inferring reward function from observed behavior.

Methods like Adam adjusting learning rates dynamically.

Learning by minimizing prediction error.

Humans assist or override autonomous behavior.

Robots learning via exploration and growth.

Inferring and aligning with human preferences.

Reusing knowledge from a source task/domain to improve learning on a target task/domain, typically via pretrained models.

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

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

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

System design where humans validate or guide model outputs, especially for high-stakes decisions.

Strategy mapping states to actions.

Combines value estimation (critic) with policy learning (actor).

Balancing learning new behaviors vs exploiting known rewards.

Continuous cycle of observation, reasoning, action, and feedback.

Loss of old knowledge when learning new tasks.

Combining simulation and real-world data.