One World Approximate Bayesian Computation (ABC) Seminar

Inspired by the "One World Probability Seminar", in April 2020 we decided to run the One World Approximate Bayesian Computation (ABC) Seminar, a (now) monthly series of seminars that will take place on Zoom on Thursdays, typically 9.30am or 1.30pm [UK time]. The idea is to gather members and disseminate results and innovation during these weeks and months under lockdown.

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

All webinars will be held on Zoom, with a link shared on the email sent via the mailing list. So if you are interested in the ABC world seminar and would like to hear from us monthly about the announced speaker, title and abstract and, most importantly, be able to join the talk, please register here.

Season 3 -- 2021/2022

Next talk: Cosma Shalizi - 23rd June 2022 , 1.30pm UK time

Title: Matching Random FeaturesLink opens in a new window
Abstract: We can do statistical inference on simulation models by adjusting the parameters in the simulation so that the values of randomly chosen functions of the simulation output match the values of those same functions calculated on the data. Results from the ”statespace reconstruction” or “geometry from a time series” literature in nonlinear dynamics indicate that just 2d + 1 such functions will typically suffice to identify a model with a d-dimensional parameter space. Results from the “random features” literature in machine learning suggest that using random functions of the data can be an efficient replacement for using optimal functions. In this talk, I sketch some of the key results, present numerical evidence about the new method’s properties, and lay out an agenda for research.
Reference: C. Shalizi. A Note on Simulation-Based Inference by Matching Random Features. ArXiv:
2111.09220, 2021.

	Speaker	Title	Abstract	Slides	Video
23.06.2022 1.30pm UK Time	Cosma ShaliziLink opens in a new window	Matching Random FeaturesLink opens in a new window	AbstractLink opens in a new window
Wednesday 25.05.2022 11.30am UK Time	Harita DellaportaLink opens in a new window	Robust Bayesian Inference for Simulator-based Models via the MMD Posterior BootstrapLink opens in a new window	Abstract	SlidesLink opens in a new window	VideoLink opens in a new window
28.04.2022	Pierre AlquierLink opens in a new window	Concentration and robustness of discrepancy-based ABC	AbstractLink opens in a new window	SlidesLink opens in a new window	VideoLink opens in a new window
31.03.2022 9.30am UK Time	David WarneLink opens in a new window	Multifidelity multilevel Monte Carlo for approximate Bayesian computationLink opens in a new window	AbstractLink opens in a new window		Recorded TalkLink opens in a new window
24.02.2022 1.30pm UK time!!	Rafael IzbickiLink opens in a new window	Likelihood-Free Frequentist Inference - Constructing Confidence Sets with Correct Conditional CoverageLink opens in a new window	AbstractLink opens in a new window		Recorded talkLink opens in a new window
03.02.2022 2.30pm UK time	Yixing WangLink opens in a new window	Posterior Collapse and Latent Variable Non-identifiabilityLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window
25.11.2021	Clara GrazianLink opens in a new window	PET-ABC: A Bayesian likelihood-free tool for kinetic modelsLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded TalkLink opens in a new window
28.10.2021	Michael GutmannLink opens in a new window	Neural Approximate Sufficient StatisticsLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded TalkLink opens in a new window
30.09.2021	Matias QuirozLink opens in a new window	Spectral Subsampling MCMC for Stationary Multivariate Time SeriesLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded TalkLink opens in a new window

Season 2 -- 2020/2021

In Season 1 and Season 2, Ritabrata Dutta, Link opens in a new window Paul Fearnhead,Link opens in a new window Gael MartinLink opens in a new windowand Judith Rousseaualso served as One World ABC Organisers.

Date	Speaker	Title	Abstract	Slides	Video
27.05.2021	Veronika RockovaLink opens in a new window	Metropolis-Hastings via ClassificationLink opens in a new window	AbstractLink opens in a new window		Recorded TalkLink opens in a new window
29.04.2021	Jakob MackeLink opens in a new window	Simulation-based inference for neuroscience (and beyond)	AbstractLink opens in a new window		Recorded talkLink opens in a new window
25.03.2021	Mijung ParkLink opens in a new window	ABCDP: Approximate Bayesian Computation with Differential PrivacyLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded TalkLink opens in a new window
10.12.2020	Matti ViholaLink opens in a new window	On the use of ABC-MCMC with inflated tolerance and post-correctionLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded TalkLink opens in a new window
12.11.2020	David NottLink opens in a new window	Marginally-calibrated deep distributional regressionLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded talkLink opens in a new window
29.10.2020	Agnieszka BorowskaLink opens in a new window	Gaussian process enhanced semi-automatic ABC for inference in a stochastic differential equation system for chemotaxisLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded talkLink opens in a new window
15.10.2020 10.30 UK time!!	David FrazierLink opens in a new window	Robust and Efficient Approximate Bayesian Computation: A Minimum Distance ApproachLink opens in a new window	AbstractLink opens in a new window	SidesLink opens in a new window	Recorded talkLink opens in a new window
01.10.2020	Marko JärvenpääLink opens in a new window	Batch simulations and uncertainty quantification in Gaussian process surrogate ABCLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded talkLink opens in a new window
17.09.2020	Flora JayLink opens in a new window and Théophile SanchezLink opens in a new window	Deep learning for population size history inference: design, comparison and combination with approximate Bayesian computationLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded talkLink opens in a new window
03.09.2020	Pierre-Alexandre MatteiLink opens in a new window and Samuel WiqvistLink opens in a new window	Partially Exchangeable Networks and Architectures for Learning Summary Statistics in Approximate Bayesian ComputationLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	Recorded talkLink opens in a new window

Season 1 -- 2020

Date	Speakers	Title	Abstract	Slides	Video
16.07.2020	Ruth BakerLink opens in a new window	Multi-fidelity Approximate Bayesian computationLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	VideoLink opens in a new window
02.07.2020	Chris DrovandiLink opens in a new window	Improving Bayesian Synthetic Likelihood via TransformationsLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	VideoLink opens in a new window
18.06.2020	Hien Duy NguyenLink opens in a new window	Approximate Bayesian computation via the energy statisticLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	VideoLink opens in a new window
04.06.2020	Irene TubikanecLink opens in a new window	Spectral density-based and measure-preserving ABC for partially observed diffusion processes. An illustration on Hamiltonian SDEsLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	VideoLink opens in a new window
21.05.2020	Gael MartinLink opens in a new window	Focused Bayesian PredictionLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	VideoLink opens in a new window
07.05.2020	Umberto PicchiniLink opens in a new window	Stratified sampling and bootstrapping for approximate Bayesian computationLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	VideoLink opens in a new window
23.04.2020	Ivis Kerama Link opens in a new windowand Richard EverittLink opens in a new window	Rare event ABC-SMC^2Link opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	VideoLink opens in a new window
09.04.2020	Dennis PrangleLink opens in a new window	Distilling importance samplingLink opens in a new window	AbstractLink opens in a new window	SlidesLink opens in a new window	VideoLink opens in a new window

Other One World Seminars

Organisers

Julyan Arbel
Richard Everitt
Michael Gutmann
Antonietta Mira
Umberto Picchini
Dennis Prangle
Christian Robert
Scott Sisson
Massimiliano Tamborrino