Where can I go for general optimization assistance?

The CVX Forum is reserved specifically for technical support for CVX and TFOCS, not for more general questions about optimization and modeling. We’re a small operation, so limiting our focus ensures that we can best help our users.

In fact, everyone is better served when you post in a forum that is properly tailored to the topic in question:

  • You. Your questions are more likely to be seen by people with the best knowledge and expertise required to answer them well.
  • This forum. It can keep its focus, its small group of active participants is not overwhelmed.
  • Other readers. Forums such as these, and those we recommend below, have an archival purpose: your question is saved for the benefit of others who might have similar questions or interests. Posting it in the appropriate forum ensures that it gains the widest audience.

If you do need help with general questions about optimization or modeling, then, here are our recommendations:

  • Math StackExchange. As the name implies, this forum covers a wide range of mathematical topics. By labeling your post with (possibly multiple) topic tags, including optimization, nonlinear programming, convex analysis, and convex optimization tags, you can help ensure that Math.SE members with the right expertise see your questions. This author is a frequent contributor there.
  • OR Exchange. This is a Q&A site sponsored by INFORMS, an international society for professionals in operations research, management science, and analytics. Not surprisingly, there is a wealth of optimization expertise on this forum.
  • CompSci StackExchange. This site is focused more on computational mathematics. If you are implementing a low-level algorithm to perform certain computations, including those involving optimization, this may be a better resource than Math.SE.
  • MathOverflow. At first, it may seem like this site is a duplicate of Math.SE. But MathOverflow’s stated focus is research-level mathematics. If you are looking for assistance to prove an obscure theorem about optimization for a journal article, this may be your resource. Otherwise, do not be surprised if your post is migrated to Math.SE or CompSci.SE.

When considering where to post a question, then, ask yourself this: is this question specific to CVX or TFOCS, or would I have the same question if I were using different software (or writing my own)? If it’s the former, then please share it here! If it’s the latter, then one of these other forums is probably a better venue.

One important point: it is considered poor etiquette to post your question on multiple forums. If you haven’t received a response after several days, or a commenter suggests that another forum would be a better fit, then you might consider posting in a different forum. But if you do, include a link to the original question, so that readers will know that it is a carefully considered duplicate.

Finally, be patient and courteous. Remember, the participants on these forums are volunteers who have their own challenges to solve. If you do not receive an answer after a couple of days, it may simply be because the participants cannot help you. Consider whether you should edit your question to provide more clarity or additional information. And never post the same question more than once on the same forum in an attempt to boost attention; it will backfire.

The Internet has become a fantastic resource for seeking help on mathematical and computational topics!


As an update to @mcg’s post, note that OR Exchange is now and likely forevermore, read only. However, Operations Research Stack Exchange is now in public Beta, is more or less the replacement for OR Exchange, and should be suitable for a large portion of optimization questions.