Difficulty: Advanced

Agents copy others’ actions.

Matrix of curvature information.

Decomposing goals into sub-tasks.

Combining simulation and real-world data.

Learning policies from expert demonstrations.

Sampling from easier distribution with reweighting.

Some agents know more than others.

Early signals disproportionately influence outcomes.

Ensuring learned behavior matches intended objective.

Measures similarity and projection between vectors.

Tendency for agents to pursue resources regardless of final goal.

Goals useful regardless of final objective.

Variable enabling causal inference despite confounding.

Computing joint angles for desired end-effector pose.

Inferring reward function from observed behavior.

Matrix of first-order derivatives for vector-valued functions.

Study of motion without considering forces.

Diffusion performed in latent space for efficiency.

Sample mean converges to expected value.

Learning without catastrophic forgetting.

Probability of data given parameters.

Mathematical foundation for ML involving vector spaces, matrices, and linear transformations.

Supplying buy/sell orders.

Estimating robot position within a map.

Eliminating variables by integrating over them.

Designing efficient marketplaces.

Effect of trades on prices.

AI discovering new compounds/materials.

Designing systems where rational agents behave as desired.

Learned subsystem that optimizes its own objective.