Matti Vihola
@mattivihola.bsky.social
Professor of Statistics, University of Jyväskylä. Computational statistics, applied probability, Monte Carlo methods, Bayesian inference.
https://iki.fi/mvihola/
https://iki.fi/mvihola/
New material on the O(T log T) coupling algorithms:
• Example where we use unbiased gradients for maximum likelihood estimation
• Generalised coupling algorithm which can handle potentials/weights that depend on current and previous state (coupling of conditional marginal particle filters)
• Example where we use unbiased gradients for maximum likelihood estimation
• Generalised coupling algorithm which can handle potentials/weights that depend on current and previous state (coupling of conditional marginal particle filters)
June 2, 2025 at 6:14 AM
New material on the O(T log T) coupling algorithms:
• Example where we use unbiased gradients for maximum likelihood estimation
• Generalised coupling algorithm which can handle potentials/weights that depend on current and previous state (coupling of conditional marginal particle filters)
• Example where we use unbiased gradients for maximum likelihood estimation
• Generalised coupling algorithm which can handle potentials/weights that depend on current and previous state (coupling of conditional marginal particle filters)
We revised our paper about particle filter forgetting: arxiv.org/abs/2309.08517. Substantial new content:
• Propagation-of-chaos bounds (TV distance between law of q particles out of N and i.i.d. draw from the ideal filter) which vanish if q = o(N)
• Method for handling out-of-sequence measurements
• Propagation-of-chaos bounds (TV distance between law of q particles out of N and i.i.d. draw from the ideal filter) which vanish if q = o(N)
• Method for handling out-of-sequence measurements
February 7, 2025 at 10:19 AM
We revised our paper about particle filter forgetting: arxiv.org/abs/2309.08517. Substantial new content:
• Propagation-of-chaos bounds (TV distance between law of q particles out of N and i.i.d. draw from the ideal filter) which vanish if q = o(N)
• Method for handling out-of-sequence measurements
• Propagation-of-chaos bounds (TV distance between law of q particles out of N and i.i.d. draw from the ideal filter) which vanish if q = o(N)
• Method for handling out-of-sequence measurements