HackerX
HomeRank #115
@goodfellow_ian

AI

@goodfellow_ian

View on X
Recommended by 2 HN usersinsight
View on X

Why Hacker News recommends it

4 comments from 2 distinct recommending readers, between 2019 and 2023.

  1. @markchen90Recommends2019
    These analogies need to be taken with a grain of salt. For instance, king - man + woman actually returns king (since woman - man ~= 0). It's only queen when you ban the words king, man, or woman (since the embeddings of queen and king are close). See Ian's tweet twitter.com/goodfellow_ian/status/11335…... or the linked paper arxiv.org/pdf/1905.09866.pdf for more information.
    Word2vec: Fish + Music = BassRead the thread
  2. @sytelusRecommends2019
    The king, queen, man, women example was in essence hand crafted and deceptive and this whole "arithmatic" property has now been debunked: twitter.com/goodfellow_ian/status/11335…...
    Why do we use word embeddings in NLP?Read the thread
  3. @nybsjytmCites it2023
    I'm assuming this is the example of "Ian Goodfellow raging about it on Twitter"? twitter.com/goodfellow_ian/status/11335…...
    NeurIPS 2023 Test of Time Winner: "Money and power certainly corrupts people "Read the thread
  4. @nybsjytmCites it2023
    I tried asking on another thread what Goodfellow rage he's referring to since all I could find was this: twitter.com/goodfellow_ian/status/11335…... If so, frankly I think it makes Mikolov sound pretty insecure.
    Word2Vec received 'strong reject' four times at ICLR2013Read the thread

More

RSS feed

Keyboard shortcuts