Here're some resources that once impressed me on the way to computer science research, collected from the web. Some15, 16 of those are also collections of resources, some13, 14 are written by Turing award winner, and some7 are suggestions from research students.

# My humble comment

[19] is a collection of experts' answers to four important questions in NLP research, namely "What do you think are the three biggest open problems in NLP at the moment?", "What would you say is the most influential work in NLP in the last decade, if you had to pick just one?", "What, if anything, has led the field in the wrong direction?" and "What advice would you give a postgraduate student in NLP starting their project now?" I can also place my humble selection here as for the first and last question.

I think the three biggest open problems in NLP at the moment are:

  • Grounded language learning
  • Language reasoning
  • Learning in low-resource settings (unsupervised, transfer, multitask, meta-learning, prior knowledge, etc)

I'm digging into the first question currently.

And I want to adopt these fourfor the suggestions to postgraduates (including me):

  • Read a lot to gain a strong background
  • Be ambitious and novel for the long-term result
  • Publish progress, even in a workshop for the short-term result
  • Spend about 10-20% time to learn to collaboration

[20] includes a series of articles subject to the aptitude of doing research written by Prof. Song-Chun Zhu. Here I put my quotation and inspiration.

There is undeniable a wide spread of utilitarianism among Chinese students and their parents, and it has become even stronger in the recent year since China has been through an economic revival. So it's more difficult and important for our Chinese students to overcome this obstacle. The only way out is to make it clear what you want and who you want to be. Some parents might explain that they just want their children to be happy and enjoy life, but as the saying goes, "the tree desires stillness but the wind will not cease". One needs to confront their fate.

Einstein once addressed a speech in Max Planck's sixtieth birthday21 in which he conveyed that there are three various motives leads people thither and dwell in the temple of science: some take to science out of a joyful sense of superior intellectual power; some others come for purely utilitarian purposes. Although they all contribute to the buildings of the temple of science, if there are only these two kinds of people, the temple would never have come to be, any more than a forest can grow which consists of nothing but creepers. Besides, there is the third kind of odd people who have a finely tempered nature longs to escape from the personal life into the world of objective perception and thought. Max Planck absolutely seat in the third group, so as many great scientists including Einstein himself. Whether the other people can become engineers, officers, tradesmen, or scientists depends on circumstances, but for the third kind of people, they mean to be scientists.

Zhu gives two analogies about two traps researchers frequently step into. One is named street lamp of research. This story is described in a book written by Michael Arbib:

It's a dark night, you see a man looking for something right under the street lamp when you walk down the street. Then you ask him: "Are you certain about your key is lost here?" "No", he replies. And you go on asking: "So why are you keep looking the key here?" "I don't know, cause here is the only bright place, where else can I find my key?"

It sounds ridiculous, but it's mostly the case we are facing today.

The second analogy called double stampede event. It comes with a story Zhu personally experienced when he was a child. The main point is an unconscious crazy research trend can tear you apart.

[13] is originally a transcription of Dr. Richard Hamming's talk at a Bell Lab seminar. It's a very special talk and is nothing about ordinary run-of-the-mill research, but great world-class research. As of where Hamming stands, he is among a few people who can carry on this kind of study and give some insight. Some points addressed in his speech does intrigue me, for example, the role luck, brain, and ambiguity about a theory play in the way to success.

Quoted from David Blackwell: "I've worked in so many areas – I'm sort of a dilettante. Basically, I'm not interested in doing research and I never have been. I'm interested in understanding, which is quite a different thing. And often to understand something you have to work it out yourself because no one else has done it."

Quoted from [12] "You can’t expect to win in the long run by somehow gaming the system or putting up false appearances."

I found Sam Altman's advice [25] on entrepreneurship also apply to CS research.

# Reference

[1] Zhihu question about Eric Xing, a professor of CMU (opens new window)

[2] Zhihu question about the status of AI possition in industry in autumn of 2019 (opens new window)

[3] Advice for Research Students - Jason Eisner @ JHU (opens new window)

[4] Applying to Ph.D. Programs in Computer Science - Mor Harchol-Balter @ CMU (opens new window)

[5] How to Be a Successful PhD Student in NLP/ML - Mark Dredze @ JHU (opens new window)

[6] Zhihu question "Should undergraduate students major in CS be encouraged to do research?", answered by Dr. Yan Gu @ CMU (opens new window)

[7] Zhihu question "As a sophomore year student, how to prepare to apply CMU?" answered by Anie @ Stanford (opens new window)

[8] Some grad school advice by Noah Smith @ UW (opens new window)

[9] Some advice for undergraduates by Noah Smith @ UW (opens new window)

[10] Advice compiled by Michael Ernst @ UW (opens new window)

[11] How to Succeed in Graduate School - Marie desJardins @ UMBC (opens new window)

[12] A Survival Guide to a PhD - Andrej Karpathy @ Stanford (opens new window)

[13] You and Your Research - Richard Hamming @ UVa (opens new window)

[14] Advice to a Beginning Graduate Student - Manuel Blum @ CMU (opens new window)

[15] Collected Advice on Research and Writing - Mark Leone @ CMU (opens new window)

[16] Grad School Advice - Jason Hong @ CMU (opens new window)

[17] What’s your advice for undergraduate student who aspires to be a research scientist in deep learning or related field one day? - Yann LeCun @ NYU (opens new window)

[18] How I Fail series - VERONIKA CHEPLYGINA @ Eindhoven University of Technology (opens new window)

[19] Frontiers in Natural Language Processing Expert Responses (opens new window)

[20] Research: Are we on the right way? - Song-Chun Zhu @ UCLA (opens new window)

[21] Principles of Research - Albert Einstein (opens new window)

[22] So You Want to Be a Research Scientist - Vincent Vanhoucke @ Google (opens new window)

[23] Advice for Researchers - Charles Sutton @ Google Brain & Edinburgh (opens new window)

[23] NLP Highlights(85) - Stress in Research, with Charles Sutton (opens new window)

[24] Interview with David Blackwell - Mathematical People (opens new window)

[25] How To Be Successful - Sam Altman (opens new window)

Last Updated: 8/31/2020, 2:30:27 PM