The inaugural SciMLCon of the SciML Open Source Software Community requests talks focused on the development and showcases of applications of the Julia-based SciML tooling. SciMLCon presentations can range from introductory to advanced, with speakers from industry and academia.

SciMLCon is focused on the advancement of open source software, in the form of tutorials and sharing expertise. To get a feel for suggested presentation styles, take a look at past programmes and recordings of JuliaCon (2021, 2020, etc.). This year’s SciMLCon will be virtual with all talks taking place in a livestream that will be live broadcast to outlets like Youtube, Twitch, and Linkedin.

We are interested in all topics that have to do with SciML, including the following incomplete list:

• Core package development (DifferentialEquations.jl, DiffEqFlux.jl, NeuralPDE.jl, etc.)
• Developments in SciML-adjacent packages (automatic differentiation, numerical linear algebra, analytics and visualization, optimization, etc.)
• Downstream package development (applications packages, PDE solvers developed on SciML packages, etc.)
• Applications of SciML software in sciences (biology, bioinformatics, health, medicine, finance, economics, physics, chemistry, etc.)
• Compiler and performance analyses, benchmarks
• Tutorials teaching and demonstrating the use of SciML software
• Discussions of software engineering and best practices for SciML use
• Benchmarking of SciML software in applications or between methods
• Use of SciML in industrial applications
• Novel scientific computing and SciML methodologies being prototyped in Julia

The most important consideration is whether the topic would be interesting to the SciML Open Source Software community.

Talks are 20 minutes long, including 5 minutes allocated for questions from the audience.

Proposal Advice

• When thinking about what makes a good topic, ask yourself what kind of presentation would make you or your friends excited.
• Talk about your own experiences. You could tell us how you use Julia in your work or for a project.
• We are interested in both innovations you developed and restrictions you encountered in Julia.
• It may be a good idea to review previous JuliaCon presentations for inspiration and ideas.
• If you are presenting about a certain project, package, or company, include links to guide the reader to more information.
• Proposals must be in clear English. Please seek help from mentors or other community members if you feel you need to!
• Try not to assume familiarity with your subject when writing the abstract; the average reader is most likely not an expert in your area of expertise.
• Avoid opaque titles; while witty titles are good, an informative title is essential. Note you will have a chance to tweak the title after being accepted.
• Avoid infomercials; while we love to hear how you use SciML in your company, SciMLCon is not the place to sell your product.
• If you want feedback before submitting a proposal or if you are interested in the mentorship program, contact us at scimlcon@julialang.org or fill out the mentorship request well ahead of the deadline. If you are interested in mentoring, please fill out the call for mentors!

If you have doubts about becoming a speaker, check out the website We are all awesome.

We look forward to your proposals and to seeing you at SciMLCon 2022!

Submission process

We are using an anonymized submissions process, to avoid selection bias related to the speaker. While enforcing double blind is impossible since most submissions should link to public open source code, all efforts are made to ensure impartial review of submissions.

Recordings and materials

We plan to video record all talks and will make them available after the conference. We do this for those who cannot attend. If you are uncomfortable with having your talk recorded, however, please contact scimlcon@julialang.org; being recorded is not a requirement to attend SciMLCon or to give a presentation. Finally, we also ask you to make your materials and recording available under a Creative Commons (default: no commercial reuse) or other open source license.

Registration for speakers

Tickets for SciMLCon 2022 are free but registration is required. We also ask that if you have the means, you donate by "buying" a ticket.

How to contact us

You can reach us with questions and concerns at scimlcon@julialang.org.

Conference Code of Conduct

SciMLCon is dedicated to providing a positive conference experience for all attendees, regardless of gender, gender identity and expression, sexual orientation, disability, physical appearance, body size, race, age, religion, or national and ethnic origin. We encourage respectful and considerate interactions between attendees and do not tolerate harassment of conference participants in any form. For example, offensive or sexual language and imagery is not appropriate for any conference venue, including formal talks and networking between sessions. Conference participants violating these standards may be sanctioned or expelled from the conference (without a refund) at the discretion of the conference organizers. Our anti-harassment policy can be found here.

Appendix

Review Guidelines and Process

For reference, below are the guidelines and processes that readers will use in reviewing your submission.

The role of reviewers is to ensure the quality of the content presented at SciMLCon.

Conflict of interest

In any case of conflict of interest, the reviewer commits to withdraw from a review and signal it to the organizers to find a replacement quickly. No reviewer should enter a review on any talk in which they are an author Or have another form of conflict of interest. See the PNAS guidelines for a definition and examples. Conflicts of interest include any work or authors with which the reviewer has "any association that poses or could be perceived as a financial or intellectual conflict of interest" (PNAS guidelines above).

Code of Conduct

The reviewer commits to reading and respecting the conference Code of Conduct in the assessment and all communications during the review process.

If a submitted abstract does not comply with the Code of Conduct, the reviewer should refer it to the organizing committee.

Criteria for the reviews

Failure to meet these criteria will result in lower scores.

1. The abstract should be easy to read, in English, and understandable for someone not working on the same topic. The title should make it easy to identify the topic of the content.

2. The abstract presented should be technically sound to the best of the reviewer's knowledge.

3. The subject should be of interest for SciMLCon, including but not limited to the topics listed on the Call for Proposal

• Core package development (DifferentialEquations.jl, DiffEqFlux.jl, NeuralPDE.jl, etc.)
• Developments in SciML-adjacent packages (automatic differentiation, numerical linear algebra, analytics and visualization, optimization, etc.)
• Downstream package development (applications packages, PDE solvers developed on SciML packages, etc.)
• Applications of SciML software in sciences (biology, bioinformatics, health, medicine, finance, economics, physics, chemistry, etc.)
• Compiler and performance analyses, benchmarks
• Tutorials teaching and demonstrating the use of SciML software
• Discussions of software engineering and best practices for SciML use
• Benchmarking of SciML software in applications or between methods
• Use of SciML in industrial applications
• Novel scientific computing and SciML methodologies being prototyped in Julia

1. Use cases of SciML in an enterprise environment are in general of interest to the conference. In particular, feedback on product development using or interacting with SciML and its ecosystem are welcome. However, talks and posters are not a suitable venue for product placement.

Scoring Criteria

The following are the criteria by which scores (1-5) should be given:

1. Applicability to the Julia community. Would users of Julia be interested in this talk for either its methods or its results? Higher scoring proposals should be of wide interest to Julia users.

2. Contributions to the community. Is this a new package for people to use? Higher scoring proposals should be code or ideas that others can use.

3. Clarity. What is the purpose of this talk? What will people learn? Higher scoring proposals should be clear as to their purpose.

4. Significance to the community. Is this something that will change the way a lot of other people use Julia or its package ecosystem? Higher scoring proposals should be more significant to SciML users. Note that this does not require scientific significance, just significance as a software or tutorial to users of SciML.

5. Topic diversity. As a community we value the diversity of applications. Proposals which are targeting new areas and fields for the SciML community to expand should be given some credit.

6. Soundness. Proposals should be technically sound. Glaring incorrectness should be highlighted and taken into account.

7. Classification. The criteria will be stricter for longer presentations.

Review Process

• Talks will be judged together in two rounds. In the first round, all reviewers will be given a selection of X many proposals to review, such that every proposal gets 3 reviews. Every review should include a score and a comment, at least a 1 sentence overview of their thoughts on the proposal. If any reviewers dropped out, we will have a second round to add more reviewers to those proposals by re-distributing to the active reviewers. After this round, the committee will ensure that each proposal has had 3 reviews.

• After the two review stages is the selection stage. The top proposals will be accepted and the bottom proposals will be rejected or re-classified and counted as a middle proposal. All proposals in the middle will then be marked for extra consideration. The amount of "top proposals" to be considered for stage 1 selection is dependent on the number of available speaker slots vs the number of proposals, and generally is reserved for average scores of 4+ without any comments of concern.

• In the second selection stage, the review committee will take the reviewer scores, comments, and re-classifications to decide the final acceptances. Rejected proposals will be discussed for re-classification before a full rejection.

• After all proposals have been classified, acceptances and rejections will be released.

Review Comments

Each review should include a comment with the score that justifies the score. For example, a comment may be like:

• "This proposal was given a 5 because it is clear, a very new application of Julia (to me), and it introduces a new package which I think many may find useful"

• "This proposal was given a 2 because, while it seems to be state-of-the-art, the closedness and cost of the software makes the talk seem only targeted at a small enterprise audience and not of general Julia user interest. If the proposal focused more on their experience using Julia rather than their product, I would have seen it as more applicable to this conference".

This Call for Papers closed on 2022-02-07 11:59 (UTC).