Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

NIPS 2014 papers

cs.stanford.edu

Post reply on HN
Print view View original

11–15 of 15 posts

Re: NIPS 2014 papers

#11

Post by jamessb »

mlla wrote: post #10
nl wrote: post #9

Earlier quoted context omitted.

Wow, those downvotes are pretty strong! Clearly I'm wrong - what am I missing?

This is done using Latent Dirichlet Allocation (LDA). The original algorithm was published by David Blei et al over ten years ago, link to the paper: http://machinelearning.wustl.edu/mlpapers/paper_files/BleiNJ... There are many machine learning libraries that have good implementations of LDA (e.g. Gensim), so it should be "relatively" straightforward to create the topics and clustering based on the abstracts of the…

I think there might be confusion about wht nl was referring to. Yes, the link is to a list (produced by Karpathy) of papers on which LDA has been performed.

But one of the listed papers is also by Kapathy ("Deep Fragment Embeddings for Bidirectional Image Sentence Mapping"), and I think this might be what nl is complimenting as being done quickly.

Re: NIPS 2014 papers

#12

Post by taneliv »

zackchase wrote: post #5

I took a stab at trying to interpret the topics output by this run of LDA. Green is one the clearest: generally convolutional deep nets, image classification, empirical work. Brown seems to have picked up on linear algebra. "Vector", "matrix", "tensor" and "decomposition" all get consistently labeled brown, as do "eigenvalues", "orthogonal" and "sparse". The rest are not as useful. Black almost always has "number", "…

Karpathy had a different interpretation (in the green bar at the top of the page). For example, purple would be neuroscience.

In addition to adjusting k, another change that might be interesting would be to include also previous years' papers in the model estimation. Changes in component (topic) weights year-over-year could perhaps reveal something about the topics, or the papers.

Re: NIPS 2014 papers

#13

Post by nl »

jamessb wrote: post #11
mlla wrote: post #10

Earlier quoted context omitted.

This is done using Latent Dirichlet Allocation (LDA). The original algorithm was published by David Blei et al over ten years ago, link to the paper: http://machinelearning.wustl.edu/mlpapers/paper_files/BleiNJ... There are many machine learning libraries that have good implementations of LDA (e.g. Gensim), so it should be "relatively" straightforward to create the topics and clustering based on the abstracts of the…

I think there might be confusion about wht nl was referring to. Yes, the link is to a list (produced by Karpathy) of papers on which LDA has been performed. But one of the listed papers is also by Kapathy ("Deep Fragment Embeddings for Bidirectional Image Sentence Mapping"), and I think this might be what nl is complimenting as being done quickly.

Yes this is the case. Thanks

Re: NIPS 2014 papers

#14

Post by redlabs4000 »

When the papers mention that code will be released, is that right now, or when the conference happens? I couldn't find any links to the code in any of the papers, including the karpathy one
Post reply on HN