So I'm doing a research project on a problem in graph theory and my conference deadline is just 3 weeks away. I need to develop a novel approximation algorithm for this problem and test it on synthetic data. I've already read a bunch of papers and have done a bunch of experiments to discover certain properties of graphs, but I haven't started writing the algorithm nor the paper. The good news is Is it possible for me to do this in 3 weeks (assuming I work a total of ~200 hours)? A deadline extension is completely out of the question. The good news is that I have an expert professor mentoring me and he'll help make the paper more mathematically rigorous. Does anyone have some advice as for how I can finish this in 3 weeks? Thanks!

closed as off-topic by Jon Custer, Scientist, scaaahu, Buzz, corey979 Oct 25 '18 at 6:19

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "The answer to this question strongly depends on individual factors such as a certain person’s preferences, a given institution’s regulations, the exact contents of your work or your personal values. Thus only someone familiar can answer this question and it cannot be generalised to apply to others. (See this discussion for more info.)" – Jon Custer, Scientist, scaaahu, Buzz, corey979
If this question can be reworded to fit the rules in the help center, please edit the question.

  • The main thing is don't skimp on sleep. You will get less done if you work over-long hours than if you get enough rest. – Patricia Shanahan Oct 24 '18 at 22:10
  • I'm still sleeping for at least 5-6 hours per night, but I'll also be working at least 10-12 hours per day. – AL1220 Oct 24 '18 at 22:23
  • 5
    Management would typically schedule a miraculous breakthrough in week 2. The real answer is to learn from this and not overpromise to such an extent in the future! – Jon Custer Oct 24 '18 at 22:58
  • 1
    That’s 5 hours gone on here... – Solar Mike Oct 25 '18 at 4:22
  • 1
    Does anyone have some advice as for how I can finish this in 3 weeks? — Sure. You can be extremely productive and extremely lucky. (If I were your advisor, I would tell you to forget about deadlines until you've actually done the work.) – JeffE Oct 25 '18 at 15:52

You won’t know if you can do it until you give it your very best shot. All we can say is that what you are asking about doing is very difficult. Some people might be able to cobble something together in 3 weeks. It won’t be as good as it might have been if they had, say, 3 months, but some rare very talented and hardworking people might produce something halfway decent.

Other people won’t come up with something good even in 3 months, or 3 years. It depends. We don’t know you, and we don’t know how hard this goal is.

I suggest that you give it your best effort. Even if you “fail” (in the narrow sense of not producing a minimum viable paper by the deadline), I believe it will be an educational and very rewarding experience to work extremely hard for the next few weeks in an attempt to meet this deadline. It will both advance your research substantially, which will yield some useful benefits in the future regardless of whether you meet this particular deadline, and teach you a lot about yourself and what you’re capable of. In my opinion, most people (especially younger, less established ones) use maybe 15% of their true intellectual capacity. Learning how to make the most of your potential is an incredibly important life skill that separates almost all successful people from everyone else, and trying to meet a crazy, impossible-looking deadline is a fantastic way to get on the path to acquiring this skill.

Finally, I don’t want to be guilty of giving you any more reasons to procrastinate by giving you long psychology blogs to read, but here is something that might give you a bit of hope/inspiration. Good luck!


I will be honest: If you expect to do real research starting from zero within 3 weeks, then you are probably on the wrong track. If that was so, then research would be easy and we would live in a world where we know and can do much more than we actually can right now. Research is difficult and takes time.

To your particular question: We of course don't know. We don't know what your concrete research project is, and we don't know how much you already have. I suspect that just implementing and debugging an algorithm you already have thought about in detail will take 3 weeks unless it is trivial. But you seem to still be in need of some of the thinking, the testing, the evaluating, the writing up results, etc.

The only person who can really help you with this is your adviser. Talk to her -- in fact, I may suggest to travel back in time and go to her about 3 months ago so you can work out what needs to be done in the time between then and now!

  • So are you saying that it takes 480 hours to do? Can we cut that time in half and get something done in 240 hours? – AL1220 Oct 24 '18 at 23:04
  • I've already conceptualized the algorithm and now I just need to get to the specifics. – AL1220 Oct 24 '18 at 23:11
  • I spoke to my advisor and he says it's possibly doable. It will probably take 25 hours to implement test the algorithm (once it's complete) and another 25 hours to evaluate the results write the research paper, so that's 50 hours. What I'm unsure of is how long it will take to do the thinking. – AL1220 Oct 24 '18 at 23:19
  • The algorithm is based on a conjecture I came up with. The conjecture (which is somewhat trivial, but quite powerful) is accurate about 90% of the time and it would be surprisingly simple to implement it, but it might not be rigorous enough. It's still a cool conjecture though. – AL1220 Oct 24 '18 at 23:24
  • How are we supposed to know whether the number of hours you quote is enough? I've implemented algorithms for which I suspect I've spent 240 hours just getting the implementation right, not counting evaluation. These were of course worth many thousands of lines of code. Maybe you have something trivial -- though usually trivial also means that someone has thought of it before. – Wolfgang Bangerth Oct 25 '18 at 1:16

Not the answer you're looking for? Browse other questions tagged or ask your own question.