Hi all!
Also, the 10th of June (AoE) is the early-bird deadline, and after that, the conference fee will increase! Also, only those who register before the 10th of June will be able to buy a StanCon 2026 Uppsala t-shirt!
We hope to see you all in Uppsala from the 17th to the 21st of August!
More information can be found here:
-– The Local Organising Committee
@mans_magnusson @hamis @avehtari
We thank our sponsors, The Beijer Foundation, eSSENCE, The Swedish Excellence Centre for Social Science and Carl Tryggers Foundation. More information about the sponsors can be found here: Sponsors – Stan Conference 2026
@mans_magnusson @hamis @avehtari Thanks to you and everyone else running StanCon this year!
Can you confirm at what rate you want postdocs and other trainees that don’t obviously fit under the student/PhD title (mentioned here) to register? In my experience, postdocs are great to have at conferences and get an immense amount out of them, but they don’t often have the grants or funds of faculty (or similar) to attend (indeed, when I was a postdoc in Canada, I took home less than my PhD student colleagues due to tax rules) so asking them to pay the full rate (which I guess does not factor in the addition 25% VAT?) may limit attendance.
I fully see your point, and I agree that getting postdocs to the conference is really important. However, there is unfortunately not much we can do about this at this point. Partly because we explicitly applied for funding to lower the fee for PhD students. We have also had the option to provide financial support and waive the fee for some of the attendees - and I know some of those were postdocs.
I think this is really important, and I have noted this down in the information to pass on to future StanCons.
Thanks for replying quickly and all your work. It is pretty disappointing to hear. I echo what @Bob_Carpenter mentioned here (and elsewhere I think but I cannot quickly find it) that we need to be thoughtful about establishing how we run StanCon and share knowledge and standards over time. Otherwise, folks like me will be far more cautious about encouraging folks to go.
Thank you for your comment.
We used the same definitions as for StanCon2024 with students/academics/industry so I’m not sure what you mean. We also try to follow all four accessibility points made in your link. I think we should always try to improve and be even more inclusive as a community, so I think how to facilitate further and even more also for postdocs is a good thing to do going forward.
A major problem with conferences that everyone concerned with equity is wrestling with is that they are another way in which the rich get richer. “Rich” academics, by which I mean ones with good grant funding, startup funds from their universities, etc., can afford to travel to Uppsala and pay for hotel and registration. This is thousands of US$ and thus beyond what a lot of people are going to be able to scare up.
I don’t know what to do about this. I like meeting in person. But then we have ridiculously good travel funding here at Flatiron Institute. We sent Charles Margossian on roughly $20K worth of travel and conference registration his last year here as a postdoc (he had papers accepted at all the major conferences and they were spread all over the world). I don’t know if he’ll have that rich of a budget as a junior faculty member at Lizzie’s university (UBC)! So it’s not just postdocs. Academic funding in the U.S. is so uncertain these days and promised money has been clawed back so that nobody knows where they stand on grant funding.
The difference between $400 and $800 registration doesn’t seem like it’s going to be a huge difference when the total cost is above $2000, but it does all add up.
Thanks @Bob_Carpenter and @mans_magnusson for this and apologies I dropped off due to some time away. Realistically the budgets offered as awards to go to conferences at universities like mine are 500-1000 CDN (or $350-700) so the 3250 SEK cost (I am including the 25% VAT) would cover just the registration. I think postdocs are different than graduate students but to lump them with faculty and expect them to pay almost double is not that thoughtful, especially if you want to include other disciplines that are less rich, and expect people to come from afar. In the end, even though we’ve been super careful about housing and plane tickets, sending two trainees to this conference is 10% of my annual budget to run a lab (pay everyone, put gas in our `mobile sampling unit’ (car), buy cocktail toothpicks to measure germinants, etc.) and so it is something I am unlikely to do again soon.
I think a broader plan to establish conference guidelines and support for Stan could help. I understand you’re following what was done last time, but I am not sure how thoughtful that was either. (And I really do not know, as I did not go last time or send anyone from my group.)
If you want to encourage postdocs and students to come, charge them both a low rate and have clear guidelines for how to apply for scholarships and when they will hear. Prioritize the scholarship budget (e.g., ahead of the gala dinner or paying anyone’s travel – I know this varies by conference and I appreciate when people say they will pay my way if I am speaker or organizer, but I also don’t mind there not being funds and then I make my decision based on that, and I don’t mind people at least asking if I have funds to cover my travel) and how to administer it.
I agree. I will note this (and link to this discussion) so next stancon will consider this more. Thank you for your suggestions and comments!
Thanks for that, and for hearing me out and I hope the meeting is great! I hope to attend another one soon.
Not just postdocs and students, but everyone. University budgets are getting cut all over, not just the US. Conferences assume everyone in industry is at rich tech company (and even they are cutting conference travel to keep everyone rowing the LLM galley), but I worked for 10 years at a mom-and-pop NLP company that was poorer than a halfway decently funded university lab and it was nearly impossible for us to pay corporate rates for conferences.
I believe the SGB has always done this and that it’s far and away the biggest part of the budget every year.
I agree, we should just keep registration rates low and work harder for sponsors. I want people like Bob of days yore from his mom-and-pop NLP to be there.
Maybe someone from the SGB could clarify and confirm this because I do not recall this from way back when, and have seen no evidence of it this time, but I suspect it has varied year to year (see here). It’s clear people are engaged on the topic, which I really appreciate. The scholarships also need to be clearly advertised, explained and decisions should go out well in advance so students are not waiting to buy plane tickets for so long. This year I don’t even recall how you found out to apply for scholarships (if it is still on the website, it is buried, maybe it is just the line `We are currently exploring options to provide discounts and stipends for junior scholars who wish to attend the conference’ on the reg page?) and decisions were announced after the early bird deadline with a special link to go back and get the early-bird rate if you did not get the scholarship.
I would agree with Lizzie’s point about making postdoc registration cheaper and I’d even be inclined to put them in the same category as students.
I can share my experience organizing StanCon '24 and some of what we tried to do (I also discussed it in this blog post…)
Scholarships
We didn’t put much effort reducing registration costs (there were the same as StanCon’23 but in pounds rather than USD, so an increase) but we tried to have a generous scholarship program. We received a lot of applicants and accepted 1/3. The scholarship fund came primarily from sponsors and ticket sails.
It’s clear to me that we could’ve gathered more sponsors, and now is probably the time to already start thinking about sponsors for future events. (Maybe we should put together a database of past sponsors for all StanCons.)
The funding from ticket sails was substantial but it came in late because many people registered late. We were rather conservative and tried not to award a new scholarship until we had funding to support the scholarship (after an early wave of awards, we started awarding on a rolling basis). But this was a dealbreaker for many applicants, because it was too late for them to book an affordable flight or apply for a visa.
There is probably a longer term strategy to be implemented where funding from the previous StanCon helps fund the next StanCon so we don’t have to wait for people to buy tickets. (Also I now have a new appreciation for why early bird registration and I’ll commit to always getting it.)
Travelling
Travelling is a challenge, and the price of flights has gone insane this year… I don’t know of a good solution to this. Some imperfect solutions:
- Record the content of the conference and make it publically available.
- Rotate the location of StanCon to make it more accessible to various communities.
Pinging the @SGB (which is just Aki, Jesse, and Teddy). I don’t think all the SGB members are on that list, so I’m also pinging @charlesm93, @SteveBroder, and @mitzimorris.
@mans_magnusson should know what the breakdown is for the upcoming StanCon.
Main costs for StanCon’24: booking rooms and campus staff, food (lunches and a potential dinner), AV team (which costs more for recordings), and scholarship. Let me see if I can find the relative proportion of each.
SGB is not doing any StanCon budgeting and all StanCon 2026 finances are handled by Uppsala University. SGB has approved who organizes StanCon 2026, but after that SGB is not part of the finances. So SGB doesn’t have anything to say on budgeting.
In StanCon 2018 Helsinki (which I organized), the registration prices were set so that academics were close to 0, students did cost money, and company tickets covered some part of student expenses. The sponsor money was used for waivers and travel support for global south and other underrepresented groups.
Finnish and Swedish postdocs tend to have good travel funding, and me and Måns forgot that there are high GDP countries where postdocs are not well paid and don’t have good travel funding. Thus we targeted waivers and travel support again for participant from Global South and other underrepresented groups (including in general people from countries with not many Bayesians and Bayesian courses). I think I’ll now remember to advise the organizers of the next StanCon to think how to improve on this. It will be hard to make it zero cost for everyone as it is unlikely that there would be so much sponsor money that everyone could get free registration and travel. @mans_magnusson can share more details on the budget, but I know that the registration fees have been planned to break even.
Yes, this was a problem also in 2018 and now in 2026. Also rolling basis is challenging if there is limited funding and there is reason to prioritize certain backgrounds.
This would make the finances more complicated. Stan has a NumFocus account which could be used in theory, but it makes things more complicated.
I have thought about this a lot, but still learning more
Thanks, Aki, Lizzie, Bob, and Charles.
Aki’s description is correct: StanCon 2026 finances are handled through Uppsala University/Academic Conferences, and the budget was planned to be very close to break-even. The latest one I saw had an expected profit of $300 on a $150 000 (ish) budget. The proportions are a little hard to see now, and it depends on how you count. But I think it is more or less the same set of main costs as at StanCon '24. I would also add travel for keynotes and the program committee.
We are actually going to work with the budget tomorrow, but what Charles mentioned is a similar problem that we also had. I think we gave around 10ish scholarships this time. We thought we could afford more, but we had to make the decision right after the early-bird period. The problem is simply that the Department of Statistics require that StanCon break even.
For future StanCons, I think we should pass on more than just informal memory. I will include this discussion and the final budget in the handover notes (now on github).
Thanks again for pushing on this. I’m sorry we did not get all of this right this year, and I hope we can make the next StanCon easier for more people to attend.
My thanks again to you @mans_magnusson and to @avehtari and @charlesm93 who have done so much to make these conferences happen. It is a herculean task in my experience! That’s especially true when the SGB does not provide a lot of support or guidance to those running the conferences.
I’d push back a little bit against this statement. The past three StanCons have all had one or two SGB members on the organization committee. I served that dual role for StanCon’23 and '24, along with Yi in '23, and I can attest that we not only acted as liaisons with the SGB but also contributed very actively to the organization. During that time, I received a ton of support and guidance from non-organizing SGB members, especially Mitzi. Furthermore, for both '23 and '24, the finance and code of conduct were handled by the SGB, with additional support from NumFocus (the non-profit that supports Stan).
I agree we can provide more support and guidance. But I think it’s less an issue of providing support from the SGB to the organization committee, and more from passing on guidance from one rotating SGB to the next and one organizing committee to the next. With more continuity, we can learn lessons from one event to the other, and for example connect more steadily with sponsors and have a more generous scholarship program.
There is and has been quite a overlap between people being members of SGB and people organizing StanCons, so it’s mostly the same people doing the volunteering work in SGB and for StanCons. In that way SGB has been involved in organizing StanCons, and asking SGB to support more StanCons is asking the involved people to support more themselves.