Jointprob: a probabilistic modeling and Bayesian statistics study community


random tree

Jointprob is a study community for probabilistic modeling and Bayesian statistics, running since August 2022.

Even though the community was initiated by Scicloj – a community for data science in Clojure – this study community is not focused on any single technology or programming language. Rather, it is open to people of diverse technical backgrounds and will continually seek dialogue and mutual inspiration across languages and technologies.

Goal

This is a long-term project aiming to create a space for Bayesian practitioners, as well as aspiring ones, to learn & explore together and support each other.

We wish to be inclusive to people of diverse backgrounds and thus create a common ground where we can broaden our perspectives and inspire each other.

Joining

The beginning of 2023 would be great timing to join, as we are starting a new reading journey.

Anyway, it is never too late to join us – we will help you to catch up.

Please use the Joining Jointprob form to join the community, to be added to our calendar events, or to update your preferences and personal information.

Also, you are always invited to write to us.

Schedule

Our current journey: Bayesian Modeling and Computation

Our past journey: Statistical Rethinking

  • 2022-07-22, initial brainstorming - event
  • 2022-08-06, prep meeting about R & Tidyverse - event (joint event with the ds4clj course)
  • 2022-08-15 - beginning of regular meetings (see the Groups below).
  • 2022-08-15/17/19/20, Session 1: Statistical Rethinking chapter 1, R4DS sections 1-4
  • 2022-08-31, 2022-09-2/3, Session 2: Statistical Rethinking chapter 2
  • 2022-09-14/16/17, Session 3: chapter 2 - review of main notions and a look into exercises
  • 2022-09-28/09-30/10-01, Session 4: chapter 3
  • 2022-10-12/14/15, Session 5: chapter 4
  • 2022-10-26/28/29, Session 6: chapter 4 - continued
  • 2022-11-9/11/12, Session 7: chapter 5
  • 2022-11-23/25/26, Session 8: chapter 6, personal regression projects
  • 2022-12-7/9/10, Session 9: more about DAGs (section 6.4), chapter 7
  • 2022-12-24, Session 10: information theory & information criteria - reporting on personal reading

Calendar events

Please refer to the Joining section to join our calendar events.

Chat

The community uses Zulip, an open-source chat platform. In some other groups we’ve been running, we’ve found it useful for chat, in-depth discussion threads, and knowledge management.

Our Zulip organization is jointprob.zulipchat.com.

You may wish to learn a little bit about the concepts of Zulip streams and topics. Note that all streams and topics (and even single messages) have URLs, that you can open at separate tabs in your browser.

It would be wonderful to present yourself at the personal intros stream, preferably as a new topic thread.

Agenda

As a community, we go through a common path, reading a book together and looking into the relevant theories, methods, and technologies. We use diverse programming languages and libraries in our practice.

In parallel to the main path of learning, we are encouraging individual and small-group projects, that participants share and present in the sessions.

At the beginning of 2023, began reading Bayesian Modeling and Computation in Python by Osvaldo A. Martin, Ravin Kumar and Junpeng Lao, possibly after a few preparation sessions for relevant background.

Previously, we read the first half of Statistical Rethinking by Richard McElreath. This book has served an important role for this community, providing a friendly and inviting path for participants of diverse backgrounds.

Sessions

We meet once in two weeks to learn together for 90 minutes. The beginning of a session is typically dedicated to following our main path of study (e.g., the current book chapter), with some hands-on exploration. Afterward, we share our progress with personal projects and explore further.

In between those regular biweekly sessions, we encourage ad-hoc sessions to explore specialized topics or practice further.

Video platform

For video meetings, we currently use Zoom. The link is shared in the calendar events.

Groups

Our regular sessions take place in groups.

All groups meet every 2 weeks and follow roughly the same reading agenda. Anybody is invited to attend one or more of the group sessions by their preference.

Here are the new groups that we set at the beginning of 2023 (in UTC time zone):

  • Group E: Wednesday 14:00-15:30 UTC
  • Group FG: Saturday 16:30-18:00 UTC

Recordings

Some parts of the sessions are recorded and shared internally in the Zulip chat. Possibly, we will aslso share some recorded parts publicly.

Projects

Participants will take on projects as individuals or in small groups.

Example projects:

  • read an article and share it with the group
  • reimplement an example we have learned using a different technology
  • explore a dataset with the methods learned

Participants are assumed to have some relevant knowledge.

To appreciate the content we are studying, it is recommended to have the following:

  • basic knowledge of probability and statistics (say, college-level intro courses)
  • programming experience in any language (a few months)
  • experience exploring data
  • an open mind

If you are not sure whether this journey fits your background, please write to us. We can think together.

Principles

The following core principles are typical of Scicloj study groups.

No experts. We do not assume that anybody is an expert in the field. We come to learn together with a student mindset.

A clear path. We will be very thoughtful about the agenda and where we wish to go. We will continually rethink and adapt our pathway going there.

Confused together. It is just fine to be confused. We will be there together and seek clarity together.

Being active. We encourage members to learn independently and take on projects. In a sense, its purpose is (also) to support those individual journeys.

Mutual curiosity. We make serious efforts to be inclusive to participants of various backgrounds. The different perspectives of our friends are part of what we wish to learn.

Contact

Please reach out: