Results for "estimation error"

Formal model linking causal mechanisms and variables.

Probability of treatment assignment given covariates.

Sum of independent variables converges to normal distribution.

Train/test environment mismatch.

Methods like Adam adjusting learning rates dynamically.

Inferring reward function from observed behavior.

Estimating robot position within a map.

Understanding objects exist when unseen.

A learning paradigm where an agent interacts with an environment and learns to choose actions to maximize cumulative reward.

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

A proper scoring rule measuring squared error of predicted probabilities for binary outcomes.

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

Reducing numeric precision of weights/activations to speed inference and reduce memory with acceptable accuracy loss.

Inputs crafted to cause model errors or unsafe behavior, often imperceptible in vision or subtle in text.

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

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

Models evaluating and improving their own outputs.

Diffusion model trained to remove noise step by step.

Two-network setup where generator fools a discriminator.

Recovering 3D structure from images.

Classical statistical time-series model.

Shift in model outputs.

Lowest possible loss.

Loss of old knowledge when learning new tasks.

Model trained on its own outputs degrades quality.

Probabilities do not reflect true correctness.

Control using real-time sensor feedback.

Performance drop when moving from simulation to reality.

RL without explicit dynamics model.

Learning policies from expert demonstrations.