Welcome to the Yelp Restaurant Photo Classification challenge! We are excited to release our first image dataset as part of this competition. Think of this challenge as a way to get to know Yelp, and your submission as a way for us to get to know you. We hope this is an interesting competition for the community - this is a problem with many possible approaches. Please reach out with any questions!
", "messages": [ { "id": "102585", "postDate": "12/23/2015 18:57:11", "content": "Welcome to the Yelp Restaurant Photo Classification challenge! We are excited to release our first image dataset as part of this competition. Think of this challenge as a way to get to know Yelp, and your submission as a way for us to get to know you. We hope this is an interesting competition for the community - this is a problem with many possible approaches. Please reach out with any questions!
", "rawMarkdown": "Welcome to the Yelp Restaurant Photo Classification challenge! We are excited to release our first image dataset as part of this competition. Think of this challenge as a way to get to know Yelp, and your submission as a way for us to get to know you. We hope this is an interesting competition for the community - this is a problem with many possible approaches. Please reach out with any questions!", "votes": null }, { "id": "102652", "postDate": "12/24/2015 06:56:12", "content": "What approaches do you suggest... aside from the obvious and expensive one of training deep networks?
", "rawMarkdown": "What approaches do you suggest... aside from the obvious and expensive one of training deep networks?", "votes": null }, { "id": "102706", "postDate": "12/24/2015 18:59:16", "content": "If you are thinking of training deep networks on single image instances with labels inherited from businesses, it might also be worth considering multi instance approaches. The wikipedia page has some resources linked (https://en.wikipedia.org/wiki/Multiple-instance_learning), and the Amores paper listed there is a very good overview; another overview is here http://www.cs.cmu.edu/~juny/MILL/review.htm. Another recent paper is http://mdenil.com/media/papers/2015-deep-multi-instance-learning.pdf.
", "rawMarkdown": "If you are thinking of training deep networks on single image instances with labels inherited from businesses, it might also be worth considering multi instance approaches. The wikipedia page has some resources linked (https://en.wikipedia.org/wiki/Multiple-instance_learning), and the Amores paper listed there is a very good overview; another overview is here http://www.cs.cmu.edu/~juny/MILL/review.htm. Another recent paper is http://mdenil.com/media/papers/2015-deep-multi-instance-learning.pdf.", "votes": null }, { "id": "102766", "postDate": "12/25/2015 18:28:49", "content": "I am an Indian citizen and I don't have any work permit for USA, will I be considered for the job if I get a good score?
", "rawMarkdown": "I am an Indian citizen and I don't have any work permit for USA, will I be considered for the job if I get a good score?", "votes": null }, { "id": "103097", "postDate": "12/28/2015 22:20:28", "content": "@Siddhartha,\nYelp does consider international candidates for roles in the US as well as in our European offices, and we sponsor US working visa. You are welcomed to attempt the challenge!
", "rawMarkdown": "Siddhartha,\r\nYelp does consider international candidates for roles in the US as well as in our European offices, and we sponsor US working visa. You are welcomed to attempt the challenge!", "votes": null }, { "id": "103714", "postDate": "01/05/2016 19:07:05", "content": "", "rawMarkdown": "", "votes": null }, { "id": "105117", "postDate": "01/20/2016 05:50:14", "content": "How many candidates are to be considered from the leader board? Is 1st place the only person to get consideration?
", "rawMarkdown": "How many candidates are to be considered from the leader board? Is 1st place the only person to get consideration?", "votes": null }, { "id": "105276", "postDate": "01/21/2016 18:03:03", "content": "@MichaelCarlon - we will consider multiple candidates, not just the first on the leaderboard.
", "rawMarkdown": "MichaelCarlon - we will consider multiple candidates, not just the first on the leaderboard.", "votes": null } ], "comments": [ { "id": 102652, "author_name": "happycube", "author_url": "", "post_date": "12/24/2015 06:56:12", "content": "What approaches do you suggest... aside from the obvious and expensive one of training deep networks?
", "votes": null, "replies": [] }, { "id": 102706, "author_name": "ody30357", "author_url": "", "post_date": "12/24/2015 18:59:16", "content": "If you are thinking of training deep networks on single image instances with labels inherited from businesses, it might also be worth considering multi instance approaches. The wikipedia page has some resources linked (https://en.wikipedia.org/wiki/Multiple-instance_learning), and the Amores paper listed there is a very good overview; another overview is here http://www.cs.cmu.edu/~juny/MILL/review.htm. Another recent paper is http://mdenil.com/media/papers/2015-deep-multi-instance-learning.pdf.
", "votes": null, "replies": [] }, { "id": 102766, "author_name": "yoyosisco", "author_url": "", "post_date": "12/25/2015 18:28:49", "content": "I am an Indian citizen and I don't have any work permit for USA, will I be considered for the job if I get a good score?
", "votes": null, "replies": [] }, { "id": 103097, "author_name": "fcchou", "author_url": "", "post_date": "12/28/2015 22:20:28", "content": "@Siddhartha,\nYelp does consider international candidates for roles in the US as well as in our European offices, and we sponsor US working visa. You are welcomed to attempt the challenge!
", "votes": null, "replies": [] }, { "id": 103714, "author_name": "roryhr", "author_url": "", "post_date": "01/05/2016 19:07:05", "content": "", "votes": null, "replies": [] }, { "id": 105117, "author_name": "lolz0r", "author_url": "", "post_date": "01/20/2016 05:50:14", "content": "How many candidates are to be considered from the leader board? Is 1st place the only person to get consideration?
", "votes": null, "replies": [] }, { "id": 105276, "author_name": "ody30357", "author_url": "", "post_date": "01/21/2016 18:03:03", "content": "@MichaelCarlon - we will consider multiple candidates, not just the first on the leaderboard.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "102585": "Welcome to the Yelp Restaurant Photo Classification challenge! We are excited to release our first image dataset as part of this competition. Think of this challenge as a way to get to know Yelp, and your submission as a way for us to get to know you. We hope this is an interesting competition for the community - this is a problem with many possible approaches. Please reach out with any questions!", "102652": "What approaches do you suggest... aside from the obvious and expensive one of training deep networks?", "102706": "If you are thinking of training deep networks on single image instances with labels inherited from businesses, it might also be worth considering multi instance approaches. The wikipedia page has some resources linked (https://en.wikipedia.org/wiki/Multiple-instance_learning), and the Amores paper listed there is a very good overview; another overview is here http://www.cs.cmu.edu/~juny/MILL/review.htm. Another recent paper is http://mdenil.com/media/papers/2015-deep-multi-instance-learning.pdf.", "102766": "I am an Indian citizen and I don't have any work permit for USA, will I be considered for the job if I get a good score?", "103097": "Siddhartha,\r\nYelp does consider international candidates for roles in the US as well as in our European offices, and we sponsor US working visa. You are welcomed to attempt the challenge!", "103714": "", "105117": "How many candidates are to be considered from the leader board? Is 1st place the only person to get consideration?", "105276": "MichaelCarlon - we will consider multiple candidates, not just the first on the leaderboard." }, "source": "meta" }