Competitive statistical inference for differential equations
Welcome to Cside 2018, the first iteration of a competitive framework comparing methods for inferring the parameters of differential equations!
In order to give competitors more time to work with the models, the conference day has been moved forward to the 26thof November.
The deadline for competitors to submit their parameter estimates has been extended from the 29th October to the 1st November.
Important: please remember to indicate your interest of participation.
Differential equations are a powerful way of providing an observed system a mathematical description. However, the act of inferring the parameters that govern the equations from noisy data, especially in a timely fashion, is challenging. The main objective of Cside is to assess what inference methods in this area are currently the state of the art.
A typical approach to assessing how well these methods perform is to generate synthetic data by solving the equations once and adding noise to reflect observational error. In this way, since the parameters that generated the data are known, one can quantify how well the method performed, subject to some assessment criteria.
However, by using simulated data we usually retain knowledge of the system that wouldn't be available in practice. This event provides researchers a way to judge the performance of methods in a blinded fashion.
Cside consists of three parts.
1. Between now and the end of October, competitors should spend some time familiarising themselves with their chosen model (see the Models Section for more details).
2. At the end of October, competitors will be given a dataset generated from their chosen model by the event organiser. You will have ten days to infer the parameters of this dataset.
3. At the end of November there will be a one day conference discussing the results of the challenge. The best 3 performing competitors for each model (9 total) will be invited to give an oral presentation on this day. A 10th competitor will be invited to give an oral presentation, based on the results of the Additional Challenge.
A timetable with the important dates of the event can be found in the Timetable Section.
There are two types of participation for the event: competitors and attendees-only. Competitors are those that will be given the competition data, will submit their inferred parameters to the conference organiser and can attend the conference day. Attendees-only are those that wish to attend the conference day, but will not be analysing the competition data.
Attendees-only are encouraged to bring a poster of their research to the conference day in order to facilitate discussion and potential collaboration with the conference delegates. Research involving systems described by differential equations is naturally multi-disciplinary and another objective of Cside is to bring together members from different backgrounds to stimulate lasting affiliations.
All research backgrounds are welcome to participate as either competitors or attendees-only.
The navigation tab on the left hand side contains the following sections
- Oral Presentations - contains the abstracts for this year's oral presentations.
- Competition Winners - the winners of Cside 2018!
- Models - contains this year's chosen models, as well as code for solving the equations.
- Assessment Criteria - details of what assessment criteria will be used to judge the performance of methods.
- Conference Day and Timetable - information of the conference day format and venue.
- Competition Timetable - a list of important dates for Cside.
- Registration of Participation - all competitors and attendees-only are asked to register their interest of participation. There are no fees charged at this stage.
- Registration of Conference Day - details of conference fees and link to full registration.
- Invitations and Prizes - information about invitation and prizes for competitors.
- Future Events - how to get in touch regarding future iterations of Cside.
For any queries, please contact Benn.Macdonald@glasgow.ac.uk
This event is