In Uniform-Cost Search (UCS), which node is always selected for expansion from the priority queue frontier?
Uniform-Cost Search (Dijkstra's search) expands the node n with the lowest path cost g(n), guaranteeing optimality for general non-negative step costs.
In a Bayesian Network, what constitutes the 'Markov Blanket' of a node X?
A node X is conditionally independent of all other nodes in the network given its Markov blanket: its parents, its children, and its children's other parents.
Which mathematical rule forms the computational foundation of the Backpropagation algorithm in Multilayer Perceptrons?
Backpropagation computes the gradient of the loss function with respect to every network weight by recursively applying the calculus chain rule backward from the output layer.
What makes OpenAI's CLIP (Contrastive Language-Image Pre-training) model capable of zero-shot image classification?
CLIP embeds both images and text captions into a shared latent space via contrastive learning; zero-shot classification is done by measuring the cosine similarity between an image embedding and prompt text embeddings.
What mathematical property defines an 'Admissible Heuristic' h(n) in A* tree search?
An admissible heuristic is optimistic, never estimating a cost greater than the actual lowest cost h*(n) to reach a goal.