Random Notes on Research

Jan 13th

Narrow Down and Go Deep

For industry, we want to solve a real problem well and completely.

But for Research, we some times focus more on go deeper in one perspective or a sub-problem of a bigger one, instead of trying to solve the whole big problem.

Since for publication, there usually exists a page limits, you cannot specify and solve all the aspects in just 10 pages. (If you can, that might mean your problem is not that difficult to solve or that specific and well-defined).

Contribution, Contribution and Contribution

There are many factors that decides whether a paper is accepted or not.

But Contributions might be one of the most important factors (or just the most important one).

Contributions can be of many things. * Raise a new important problem * Formulate an old problem with a useful new form which may be much more easier to analyze * Solve an already defined problem with new and better method * Did great work in boosting the development of a field of study * To be add more…

Jan 16th

  • new techniques in related fields may …

Some quotes from my supervisor (Oct. 12)

  • The very nature of research poses its own special risk. In research, we daily face the uncertainty of whether our chosen approach will succeed or fail. We struggle to unravel very complex puzzles, often making no visible progress for weeks or months, sometimes for years… The special risk of research starts with the high probability that any particular attempt will fail and follows from the resulting experience of repeated failure. Research carries a special risk of discouragement. - From “Technology and Courage”, Ivan Sutherland (Turing Award).

  • Don’t make the same decision twice. Spend time and thought to make a solid decision the first time so that you don’t revisit the issue unnecessarily. If you’re too willing to reopen issues, it interferes not only with your execution but only with your motivation to make a decision in the first place. - Bill Gates

  • 在行动上尽可能坚定果断,一旦选定某种看法,哪怕它十分可疑,也毫不动摇地坚决遵循,就像它十分可靠一样。这样做是效法森林里迷路的旅客,他们决不能胡乱地东走走西撞撞,也不能停在一个地方不动,必须始终朝着一个方向尽可能笔直地前进,尽管这个方向在开始的时候只是偶然选定的,也不要由于细小的理由改变方向,因为这样做即便不能恰好走到目的地,至少最后可以走到一个地方,总比困在树林里面强。为人处世也是这样,我们的行动常常必须当机立断,刻不容缓。有一条非常可靠的真理,就是在无法分明分辨哪种看法最正确的时候必须遵从或然性最大的看法。 - 笛卡尔

Nov 1st

  1. To have a taxonomy, it is better to categorize the work based on different tasks.