Results for "dual formulation"

Alternative formulation providing bounds.

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

Combination of cooperation and competition.

Modeling interactions with environment.

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

Attention where queries/keys/values come from the same sequence, enabling token-to-token interactions.

Architecture based on self-attention and feedforward layers; foundation of modern LLMs and many multimodal models.

Feature attribution method grounded in cooperative game theory for explaining predictions in tabular settings.

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

Allows gradients to bypass layers, enabling very deep networks.

Empirical laws linking model size, data, compute to performance.

Separates planning from execution in agent architectures.

Expected cumulative reward from a state or state-action pair.

Autoencoder using probabilistic latent variables and KL regularization.

Simultaneous Localization and Mapping for robotics.

Agent reasoning about future outcomes.

Scaling law optimizing compute vs data.

Average value under a distribution.

Belief before observing data.

Converts constrained problem to unconstrained form.

Equations governing how system states change over time.

Ensuring models comply with lending fairness laws.

AI proposing scientific hypotheses.

Designing efficient marketplaces.

Tradeoff between safety and performance.