WingNews logo WingNews
top | item 30172356

(no title)

hapanin | 4 years ago

I want collaborate on publishable ML research.

I'm currently working as a machine learning engineer, but doing a lot more engineering than learning. I'm also applying to PhD programs with focuses in knowledge graphs and common sense reasoning, but willing to learn any other field. If I don't get in then some independent research is still a good career move.

I have ideas of my own (goal oriented question asking) but happy to run support on other peoples stuff.

discuss

order

fxtentacle|4 years ago

Most stuff doesn't get published because preparing things for publication is such a grind.

1. I have a complete optical flow training data set here. It is built specifically to highlight the shortcomings of some state of the art AI algorithms and (as expected) they fail on it. We used it internally to fix those issues on our own proprietary AI. To turn this dataset into a publication, someone would now need to get access to or re-implement most of the top algorithms on other benchmarks like Sintel and run them against the new dataset. Otherwise, there would be no comparison scores and, hence, no use for someone to evaluate their algorithm on the new data set.

2. By participating in the gocoder Bomberland AI competition, I noticed that most state of the art RL AI algorithms fail badly in an environment with a non-deterministic enemy. It would probably be very useful to package that environment as a OpenAI gym python package and then do a proper "best out of 3" evaluation of common algorithms on that environment. It'll mostly be failures, which is a great backdrop against which to propose a small tweak to DQN that makes things work, which is to estimate the variance in addition to the mean so that you can work on the 30% percentile of the expectation of the state action score.

Both dataset papers and AI algorithm ranking papers tend to get cited a lot. But they require a lot of effort to produce.

hapanin|4 years ago

Those both actually sound quite interesting to me. What is your ideal collaboration setup / level of involvement with these? I'll start reading up. Please shoot me an email (see profile).

dummyvariable|4 years ago

Are you planning on doing a part time PhD? I am interested in the possibility, although I do not have a ML background.

hapanin|4 years ago

I would do PhD full time since my current job is not for anything publishable or in my intended field of study.

vaibkv|4 years ago

I'd like to collaborate on this too. lmk if you have any cool ideas.

hapanin|4 years ago

Just the one I wrote my applications on: a method for training a model to ask clarifying or goal oriented questions using QA-GNN. My email is in my profile, we can talk there.