Relations between machine learning problems



Home

Description

Deadlines

Organisers

Format

Submissions

Contact

Call for Submissions


SUBMISSION DEADLINE EXTENDED TO 16 October

(To align with a number of other workshop deadlines, we have extended the submission deadline to 16 October. We will send out decisions before the early registration deadline)
Submissions are invited to the workshop. The selection  criteria we will use are:
  • Alignment: Does the proposal address the core topic of the workshop, or is it a force fit?  Does the submission relate problems, or is it solely focussed on particular techniques?
  • Clarity, novelty and significance of the results
  • Accessibility and potential interest to the audience (so if you wish to use exotic mathematics that is rarely used in machine learning, you will need to make the submission adequately self-contained)
  • Rigor: if there are theorems claimed, you need proofs; if your work rests on experiments, they should be done well; if the submission is more of a position statement, it needs to be closely argued and well situated relative to the literature.
  • Diversity and balance of the program (all other things being equal, we would like a rich program, balancing theory and practice, as well as many different approaches). We would like to be surprised and fascinated by your submission!
 We welcome submissions for both short (15 minutes)  and longer presentations (30 minutes).
 You should indicate in your submission whether you are proposing a short or a long talk.

Format

There is no fixed format for submission. However you need to provide enough detail for us to be able to evaluate your work. Thus a one page summary alone will not suffice. Draft papers that are in preparation for other venues are fine. (We do not propose to publish proceedings so there is no problem in overlap).

Work that has already been published or accepted will be considered, but you must declare its status at submission and we would expect only published work that is particularly pertinent and high quality would be considered for presentation at the workshop.

If you do submit long full papers, it should be very clear (from the abstract and introduction) what is the  key contribution and what  is the connection to the workshop. You may wish to consider also including a one page summary that makes this explicit.

Please provide your submission in pdf format.

Conditions

We expect that all presenters will provide their slides immediately prior to the workshop and allow us to include them on the workshop page. If there are associated papers we would also expect to be able to link to them (preprints are fine).

Submissions will be reviewed in confidence by the organisers, who may seek third party opinions in some cases.

Contact

Email submissions to rml@anu.edu.au by 16 October