0

Problem

I am now working on a relatively new field in machine learning. My advisor suggested me to read milestone papers in this field. I think The first paper that proposed the problem is surely an important one. However, I do not know how to find it even after reading several review papers (actually none of them mentions the origin of the field).

Though finding the origin of the field may not sound useful, it is important to me since I have some fundamental doubts about the problem description, etc.

Could anyone help me, thank you in advance.

Improve this question
2
  • 1
    It's not clear to me how recent and how specific is this new field in ML, but in case it's significant enough this Wikipedia timeline might help: en.wikipedia.org/wiki/Timeline_of_machine_learning
    – Erwan
    Jan 20, 2019 at 20:55
  • 1
    It is quote possible that there is no "first paper". Some topics and/or techniques and/or fields evolve organically from their predecessors over many years.
    – JeffE
    Jan 21, 2019 at 5:18

1 Answer 1

Reset to default
1

  1. THe Wikipedia article on this topic is reasonably good and has some references. I would look at that article and the references within.

  2. Try tracing back "references of references" of reviews or important papers. (You don't have to read in depth.) But you may figure out the better papers. May also find those that have a perspective on (3).

  3. In general, I think "machine learning" is a misnomer and a buzzword. Talking about it as some organic "thing" is not correct. You should see it as a bit of a fad (especially the use of the term) and within the broader fields of statistics, algorithms, optimization, etc. It's not like discovering X-rays or something.

Improve this answer

Not the answer you're looking for? Browse other questions tagged .