One World Approximate Bayesian Computation (ABC) Seminar
The focus of this seminar series is not limited to ABC. Indeed, it spans over likelihood-free inference, ML related techniques, Approximate Bayesian Inference, etc. Feel free to contact any of the organisers if you want to suggest yourself or someone else for a talk.
How it works and mailing list
Season 4 -- 2022/2023
2nd February, 1.30pm UK time: Joel Dyer (University of Oxford)
Title: Approximate Bayesian Computation with Path Signatures
Abstract: Simulation models often lack tractable likelihood functions, making likelihood-free inference methods indispensable. Approximate Bayesian computation (ABC) generates likelihood-free posterior samples by comparing simulated and observed data through some distance measure, but existing approaches are often poorly suited to time series simulators, for example due to an independent and identically distributed data assumption. In this talk, we will discuss our work on the use of path signatures in ABC as a means to handling the sequential nature of time series data of different kinds. We will begin by discussing popular approaches to ABC and how they may be extended to time series simulators. We will then introduce path signatures, and discuss how signatures naturally lead to two instances of ABC for time series simulators. Finally, we will demonstrate that the resulting signature-based ABC procedures can produce competitive Bayesian parameter inference for simulators generating univariate, multivariate, irregularly spaced, and even non-Euclidean sequences.
Reference: J. Dyer, P. Cannon, S. M Schmon (2022). Approximate Bayesian Computation with Path Signatures. arXiv preprint 2106.12555Link opens in a new window
Speaker | Title | Abstract | Slides | Video | |
02.02.2023 1.30pm |
Joel DyerLink opens in a new window | Approximate Bayesian Computation with Path SignaturesLink opens in a new window | AbstractLink opens in a new window | ||
24.11.2022 |
Yuexi WangLink opens in a new window | Adversarial Bayesian SimulationLink opens in a new window | AbstractLink opens in a new window | SlidesLink opens in a new window | VideoLink opens in a new window |
20.10.2022, 2.30pm UK time |
Radu CraiuLink opens in a new window | Bayesian Computation Strategies for Big Data and Intractable Models | AbstractLink opens in a new window | SlidesLink opens in a new window | VideoLink opens in a new window |
29.09.2022, 1.30pm Uk time |
Till HoffmannLink opens in a new window | Minimizing the Expected Posterior Entropy Yields Optimal Summary StatisticsLink opens in a new window | AbstractLink opens in a new window | SlidesLink opens in a new window | VideoLink opens in a new window |
Season 3 -- 2021/2022
In Seasons 1-3, Julyan ArbelLink opens in a new window and Scott SissonLink opens in a new window also served as One World ABC Organisers.
Season 2 -- 2020/2021
In Season 1 and Season 2, Ritabrata Dutta, Link opens in a new windowPaul Fearnhead,Link opens in a new windowGael MartinLink opens in a new windowand Judith Rousseaualso served as One World ABC Organisers.
Season 1 -- 2020
In Season 1 and Season 2, Ritabrata Dutta, Link opens in a new windowPaul Fearnhead,Link opens in a new windowGael MartinLink opens in a new windowand Judith Rousseaualso served as One World ABC Organisers.
Other One World Seminars
- One World Probability SeminarLink opens in a new window
- One World PDE SeminarLink opens in a new window
- Waves in One WorldLink opens in a new window
- One World Mathematical Game Theory SeminarLink opens in a new window
- One World Seminar: Mathematical Methods for Arbitrary Data SourcesLink opens in a new window
- One World Cognitive Psychology SeminarLink opens in a new window
- One World IMAGing and INvErse problems (IMAGINE) SeminarLink opens in a new window
- One World Optimization SeminarLink opens in a new window
- One World Actuarial Research SeminarLink opens in a new window
- One World Mathematics of INformation, Data, and Signals (MINDS) SeminarLink opens in a new window
-
One World Stochastic Numerics and Inverse Problems (SNIP) SeminarLink opens in a new window
-
Bachelier Finance Society One World SeminarLink opens in a new window
-
One World Mathematical Physics SeminarLink opens in a new window
-
One World seminar on combinatorics on wordsLink opens in a new window
-
One World Seminar Series on the Mathematics of Machine LearningLink opens in a new window
Organisers
Richard EverittLink opens in a new window
Michael GutmannLink opens in a new window
Antonietta MiraLink opens in a new window
Umberto PicchiniLink opens in a new window
Dennis PrangleLink opens in a new window
Christian RobertLink opens in a new window
Massimiliano TamborrinoLink opens in a new window