On the Efficient Marginalization of Probabilistic Sequence Models

arXiv:2403.04005v1 Announce Type: cross
Abstract: Real-world data often exhibits sequential dependence, across diverse domains such as human behavior, medicine, finance, and climate modeling. Probabilistic methods capture the inherent uncertainty associated with prediction in these contexts, with autoregressive models being especially prominent. This dissertation focuses on using autoregressive models to answer complex probabilistic queries that go beyond single-step prediction, such as the timing of future events or the likelihood of a specific event occurring before another. In particular, we develop a …

abstract arxiv autoregressive models behavior climate climate modeling cs.lg data diverse domains finance human medicine modeling prediction stat.ml type uncertainty world