Some notes

These are rough notes written primarily for learning and self-studying. They are organized as a wiki, following the Zettelkasten method of research and knowledge management.

Please drop me a message if you find a mistake!

• Latent neural process (LNP): Derivation of an ELBO-like objective
• Natural products: Chemicals produced by living organisms
• Cantor's diagonal argument: There are more real than natural numbers
• Monte Carlo gradient estimator: Taking the gradient of an expectation
• Score function estimator: A Monte Carlo gradient estimator
• Reparameterization trick: A Monte Carlo gradient estimator
• Monte Carlo approximation: Estimate an expectation with samples
• Gated recurrent unit (GRU): At each step, decide if each element of context vector is replaced
• Random variable: A map from a probability space to a measurable space
• Probability space: A mathematical construct to model a random experiment
• Irreducible Markov chain: The whole state space is a single recurrence class
• Recurrence in Markov chains: State space is split into recurrent and transient regions