I think many people have played the game werewolf, the people need to figure out who is a werewolf from villagers by analyzing the information provided by everyone.
\n\nhttps://www.playwerewolf.co/rules
\n\nAlthough I am not an expert on this game, I know one key point in the game is to observe how every player defense themselves.
\n\nIn many occasions, players only have part information so some times they can just figure out someone may be a werewolf. However, if that person is not a werewolf. he or she usually defense himself very hard, trying to prove he is just a villager.
\n\nHowever, if that person is a werewolf, usually they focusing on how to pretend to be innocent instead of focusing on defending themselves very hard.
\n\nSo right now, peoples found some possible werewolves in this competition, but we didn't see any convincing defense from them. That is interesting.
\n\n> n01z3 wrote\n> \n> > All your comments look very emotional and not very constructive. I do not want to waste time on this kind of discussion.\n> All I can say about public results: seems that I overfitted. I have done this many times and not always successfully. TGS Salt: 11th (in Gold) Public -> 40th Private; Airbus: 2nd (!) Public -> 36th Private.\n> This time, more fortunate. And in the iMaterialist Fashion competition I unintentionally acted without local validation at all and the public LB correlated very well with the private LB.
\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/95223#550031
\n\n> ValeriyBabushkin wrote\n> \n> >Nothing new for me\nIn this competition, I dropped from 3-rd to 129
\n\n> >And here from 2 to 128\n> >https://mlbootcamp.ru/round/12/rating/?results_filter=preliminary \n> > https://mlbootcamp.ru/round/12/rating/?results_filter=final
\n\n> >129 - 3 = 128 - 2 (coincedence, ins't it ?)
\n\n> >On Airbus from 7 to 47\n> >https://www.kaggle.com/c/airbus-ship-detection/leaderboard
\n\n> >Here from 1 st to 35\n> >https://mlbootcamp.ru/round/11/rating/?results_filter=preliminary\n> >https://mlbootcamp.ru/round/11/rating/?results_filter=final
\n\n> >That is basically the reason why I prefer to wait until the final results are revealed
\n\n> >How I do this?\n> >Each time is different. Sometimes it is 5-6 layers stacking, sometimes it is a false correlation with the data, sometimes it is a surplus of computing power in terms of DGX
\n\n> >But basically, that I was speaking about the leak, I was thinking about 0.7 + Each time I see someone so high - I am thinking about the leak
\n\n> ValeriyBabushkin wrote\n>
\n\n\n\n\n\n\nI would assemble everything into a working solution if I have to provide it, however, I am not in the medals and don't have to
\n
> ValeriyBabushkin wrote\n>
\n\n\n\n\n\n\ndon't think place outside medals could be reviewed as strong
\n
https://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#549863\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#549901\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#550001
", "messages": [ { "id": 550346, "postDate": "2019-06-11T14:21:16.710Z", "content": "I think many people have played the game werewolf, the people need to figure out who is a werewolf from villagers by analyzing the information provided by everyone.
\n\nhttps://www.playwerewolf.co/rules
\n\nAlthough I am not an expert on this game, I know one key point in the game is to observe how every player defense themselves.
\n\nIn many occasions, players only have part information so some times they can just figure out someone may be a werewolf. However, if that person is not a werewolf. he or she usually defense himself very hard, trying to prove he is just a villager.
\n\nHowever, if that person is a werewolf, usually they focusing on how to pretend to be innocent instead of focusing on defending themselves very hard.
\n\nSo right now, peoples found some possible werewolves in this competition, but we didn't see any convincing defense from them. That is interesting.
\n\n> n01z3 wrote\n> \n> > All your comments look very emotional and not very constructive. I do not want to waste time on this kind of discussion.\n> All I can say about public results: seems that I overfitted. I have done this many times and not always successfully. TGS Salt: 11th (in Gold) Public -> 40th Private; Airbus: 2nd (!) Public -> 36th Private.\n> This time, more fortunate. And in the iMaterialist Fashion competition I unintentionally acted without local validation at all and the public LB correlated very well with the private LB.
\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/95223#550031
\n\n> ValeriyBabushkin wrote\n> \n> >Nothing new for me\nIn this competition, I dropped from 3-rd to 129
\n\n> >And here from 2 to 128\n> >https://mlbootcamp.ru/round/12/rating/?results_filter=preliminary \n> > https://mlbootcamp.ru/round/12/rating/?results_filter=final
\n\n> >129 - 3 = 128 - 2 (coincedence, ins't it ?)
\n\n> >On Airbus from 7 to 47\n> >https://www.kaggle.com/c/airbus-ship-detection/leaderboard
\n\n> >Here from 1 st to 35\n> >https://mlbootcamp.ru/round/11/rating/?results_filter=preliminary\n> >https://mlbootcamp.ru/round/11/rating/?results_filter=final
\n\n> >That is basically the reason why I prefer to wait until the final results are revealed
\n\n> >How I do this?\n> >Each time is different. Sometimes it is 5-6 layers stacking, sometimes it is a false correlation with the data, sometimes it is a surplus of computing power in terms of DGX
\n\n> >But basically, that I was speaking about the leak, I was thinking about 0.7 + Each time I see someone so high - I am thinking about the leak
\n\n> ValeriyBabushkin wrote\n>
\n\n\n\n\n\n\nI would assemble everything into a working solution if I have to provide it, however, I am not in the medals and don't have to
\n
> ValeriyBabushkin wrote\n>
\n\n\n\n\n\n\ndon't think place outside medals could be reviewed as strong
\n
https://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#549863\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#549901\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#550001
", "rawMarkdown": "I think many people have played the game werewolf, the people need to figure out who is a werewolf from villagers by analyzing the information provided by everyone.\n\nhttps://www.playwerewolf.co/rules\n\nAlthough I am not an expert on this game, I know one key point in the game is to observe how every player defense themselves.\n\nIn many occasions, players only have part information so some times they can just figure out someone may be a werewolf. However, if that person is not a werewolf. he or she usually defense himself very hard, trying to prove he is just a villager.\n\nHowever, if that person is a werewolf, usually they focusing on how to pretend to be innocent instead of focusing on defending themselves very hard. \n\nSo right now, peoples found some possible werewolves in this competition, but we didn't see any convincing defense from them. That is interesting.\n\n> **n01z3 wrote**\n> \n> > All your comments look very emotional and not very constructive. I do not want to waste time on this kind of discussion.\n> All I can say about public results: seems that I overfitted. I have done this many times and not always successfully. TGS Salt: 11th (in Gold) Public -> 40th Private; Airbus: 2nd (!) Public -> 36th Private.\n> This time, more fortunate. And in the iMaterialist Fashion competition I unintentionally acted without local validation at all and the public LB correlated very well with the private LB.\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/95223#550031\n\n> **ValeriyBabushkin wrote**\n> \n> >Nothing new for me\nIn this competition, I dropped from 3-rd to 129\n\n> >And here from 2 to 128\n> >https://mlbootcamp.ru/round/12/rating/?results_filter=preliminary \n> > https://mlbootcamp.ru/round/12/rating/?results_filter=final\n\n> >129 - 3 = 128 - 2 (coincedence, ins't it ?)\n\n> >On Airbus from 7 to 47\n> >https://www.kaggle.com/c/airbus-ship-detection/leaderboard\n\n> >Here from 1 st to 35\n> >https://mlbootcamp.ru/round/11/rating/?results_filter=preliminary\n> >https://mlbootcamp.ru/round/11/rating/?results_filter=final\n\n> >That is basically the reason why I prefer to wait until the final results are revealed\n\n> >How I do this?\n> >Each time is different. Sometimes it is 5-6 layers stacking, sometimes it is a false correlation with the data, sometimes it is a surplus of computing power in terms of DGX\n\n> >But basically, that I was speaking about the leak, I was thinking about 0.7 + Each time I see someone so high - I am thinking about the leak\n\n> **ValeriyBabushkin wrote**\n> \n>>I would assemble everything into a working solution if I have to provide it, however, I am not in the medals and don't have to\n\n> **ValeriyBabushkin wrote**\n> \n>>don't think place outside medals could be reviewed as strong\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#549863\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#549901\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#550001", "votes": 16 }, { "id": 550709, "postDate": "2019-06-12T00:36:40.413Z", "content": "I feel very disappointed about the platform, they release the medals without any explanation of the private sharing things.
", "rawMarkdown": "I feel very disappointed about the platform, they release the medals without any explanation of the private sharing things. ", "votes": 7 }, { "id": 550700, "postDate": "2019-06-12T00:02:26.390Z", "content": "All the teams did not drop out places except X5.
\n\nAll the teams did not lost points compare with public except X5.
\n\nX5 is outliers in this competition
\n\nWhat X5 did was an insult to statistics.
\n\nIt's hard to imagine that the GM/M in X5 have never learned statistics.
", "rawMarkdown": "All the teams did not drop out places except X5.\n\nAll the teams did not lost points compare with public except X5.\n\nX5 is outliers in this competition\n\nWhat X5 did was an insult to statistics.\n\nIt's hard to imagine that the GM/M in X5 have never learned statistics.", "votes": 4, "replies": [ { "id": 550702, "postDate": "2019-06-12T00:08:00.160Z", "content": "cheating is allowed in kaggle :)
", "rawMarkdown": "cheating is allowed in kaggle :)", "votes": 5 }, { "id": 550737, "postDate": "2019-06-12T01:27:19.477Z", "content": "X5 did it! Keep cheating！
", "rawMarkdown": "X5 did it! Keep cheating！", "votes": 4 }, { "id": 550750, "postDate": "2019-06-12T02:10:40.323Z", "content": "呜呜呜
", "rawMarkdown": "呜呜呜", "votes": 2 }, { "id": 551729, "postDate": "2019-06-13T03:55:09.350Z", "content": "Don't cry, Yanzu Wu.
", "rawMarkdown": "Don't cry, Yanzu Wu. ", "votes": 1 } ] }, { "id": 550665, "postDate": "2019-06-11T22:24:36.763Z", "content": "Really constructive post :)
\n\n\n\n\nI know one key point in the game is to observe how every player defense themselves.
\n \nIn many occasions, ...
\n
And I played this game and not every player defends herself like you describe. So what?..) Considering your example, I think you should agree that a lot of clever people understand this logic you suggested and do everything conversely. Why not? So this is very subjective)
\n\nAbout this whole situation - not to you personally but for everyone literally exploding (sometimes even with extremely dirty talks from their mouths) who probably will read it - really guys? So if you bomb forums, presenting yourself as very offended people, you really think you are smart or what?) This whole situation looks terribly similar with what happens in the world more and more, when politicians blame the others without any evidence. Now - OK, but how a normal person would buy it then? HOW CAN YOU SERIOUSLY BLAME A PERSON WITHOUT A PROOF WOULD BE FOUND IN A FEW DAYS IF IT EXISTS? I agree that these people blamed at the forum act strange - but these are only GUESSES, not PROOFS. Moving back to our politicians - so either they are dumb, or they are smart and do this to discredit someone. But you all are not politicians, right? So on which side are you, 1 or 2?)) Think about it - if Kaggle officials agreed to take a look at the results, maybe we should wait for the results? If not, then write to them and notify about it) Otherwise your shoot at the discussions is pathetic and makes you look miserable, probably you have to drop it in some other well-known place))
\n\np.s. meow meow meo meow
", "rawMarkdown": "Really constructive post :)\n\n> I know one key point in the game is to observe how every player defense themselves.\n\n> In many occasions, ...\n\nAnd I played this game and not every player defends herself like you describe. So what?..) Considering your example, I think you should agree that a lot of clever people understand this logic you suggested and do everything conversely. Why not? So this is very subjective)\n\nAbout this whole situation - not to you personally but for everyone literally exploding (sometimes even with extremely dirty talks from their mouths) who probably will read it - really guys? So if you bomb forums, presenting yourself as very offended people, you really think you are smart or what?) This whole situation looks terribly similar with what happens in the world more and more, when politicians blame the others without any evidence. Now - OK, but how a normal person would buy it then? HOW CAN YOU SERIOUSLY BLAME A PERSON WITHOUT A PROOF WOULD BE FOUND IN A FEW DAYS IF IT EXISTS? I agree that these people blamed at the forum act strange - but these are only GUESSES, not PROOFS. Moving back to our politicians - so either they are dumb, or they are smart and do this to discredit someone. But you all are not politicians, right? So on which side are you, 1 or 2?)) Think about it - if Kaggle officials agreed to take a look at the results, maybe we should wait for the results? If not, then write to them and notify about it) Otherwise your *shoot* at the discussions is pathetic and makes you look miserable, probably you have to drop it in some other well-known place))\n\np.s. meow meow meo meow", "replies": [ { "id": 550704, "postDate": "2019-06-12T00:22:23.583Z", "content": "So what I am blaming here?
", "rawMarkdown": "So what I am blaming here?" }, { "id": 551923, "postDate": "2019-06-13T09:40:41.193Z", "rawMarkdown": "", "votes": -4, "isDeleted": true } ] }, { "id": 550374, "postDate": "2019-06-11T15:00:09.540Z", "content": "Thanks for summarizing it. But don’t forget that werewolf board game is based on a russian game called Mafia
", "rawMarkdown": "Thanks for summarizing it. But don’t forget that werewolf board game is based on a russian game called Mafia ", "votes": -1 } ], "comments": [ { "id": 550709, "author_name": "Yiheng Wang", "author_url": "", "post_date": "2019-06-12T00:36:40.413000", "content": "I feel very disappointed about the platform, they release the medals without any explanation of the private sharing things.
", "votes": 7, "replies": [] }, { "id": 550700, "author_name": "30CrMnSiA", "author_url": "", "post_date": "2019-06-12T00:02:26.390000", "content": "All the teams did not drop out places except X5.
\n\nAll the teams did not lost points compare with public except X5.
\n\nX5 is outliers in this competition
\n\nWhat X5 did was an insult to statistics.
\n\nIt's hard to imagine that the GM/M in X5 have never learned statistics.
", "votes": 4, "replies": [ { "id": 550702, "author_name": "earhian", "author_url": "", "post_date": "2019-06-12T00:08:00.160000", "content": "cheating is allowed in kaggle :)
", "votes": 5, "replies": [] }, { "id": 550737, "author_name": "SeuTao", "author_url": "", "post_date": "2019-06-12T01:27:19.477000", "content": "X5 did it! Keep cheating！
", "votes": 4, "replies": [] }, { "id": 550750, "author_name": "JIANJIAN", "author_url": "", "post_date": "2019-06-12T02:10:40.323000", "content": "呜呜呜
", "votes": 2, "replies": [] }, { "id": 551729, "author_name": "seefun", "author_url": "", "post_date": "2019-06-13T03:55:09.350000", "content": "Don't cry, Yanzu Wu.
", "votes": 1, "replies": [] } ] }, { "id": 550665, "author_name": "Evgeny Kovalev", "author_url": "", "post_date": "2019-06-11T22:24:36.763000", "content": "Really constructive post :)
\n\n\n\n\nI know one key point in the game is to observe how every player defense themselves.
\n \nIn many occasions, ...
\n
And I played this game and not every player defends herself like you describe. So what?..) Considering your example, I think you should agree that a lot of clever people understand this logic you suggested and do everything conversely. Why not? So this is very subjective)
\n\nAbout this whole situation - not to you personally but for everyone literally exploding (sometimes even with extremely dirty talks from their mouths) who probably will read it - really guys? So if you bomb forums, presenting yourself as very offended people, you really think you are smart or what?) This whole situation looks terribly similar with what happens in the world more and more, when politicians blame the others without any evidence. Now - OK, but how a normal person would buy it then? HOW CAN YOU SERIOUSLY BLAME A PERSON WITHOUT A PROOF WOULD BE FOUND IN A FEW DAYS IF IT EXISTS? I agree that these people blamed at the forum act strange - but these are only GUESSES, not PROOFS. Moving back to our politicians - so either they are dumb, or they are smart and do this to discredit someone. But you all are not politicians, right? So on which side are you, 1 or 2?)) Think about it - if Kaggle officials agreed to take a look at the results, maybe we should wait for the results? If not, then write to them and notify about it) Otherwise your shoot at the discussions is pathetic and makes you look miserable, probably you have to drop it in some other well-known place))
\n\np.s. meow meow meo meow
", "votes": 0, "replies": [ { "id": 550704, "author_name": "Strideradu", "author_url": "", "post_date": "2019-06-12T00:22:23.583000", "content": "So what I am blaming here?
", "votes": 0, "replies": [] }, { "id": 551923, "author_name": "", "author_url": "", "post_date": "2019-06-13T09:40:41.193000", "content": "", "votes": -4, "replies": [] } ] }, { "id": 550374, "author_name": "brumberg", "author_url": "", "post_date": "2019-06-11T15:00:09.540000", "content": "Thanks for summarizing it. But don’t forget that werewolf board game is based on a russian game called Mafia
", "votes": -1, "replies": [] } ], "raw_markdown_by_id": { "550346": "I think many people have played the game werewolf, the people need to figure out who is a werewolf from villagers by analyzing the information provided by everyone.\n\nhttps://www.playwerewolf.co/rules\n\nAlthough I am not an expert on this game, I know one key point in the game is to observe how every player defense themselves.\n\nIn many occasions, players only have part information so some times they can just figure out someone may be a werewolf. However, if that person is not a werewolf. he or she usually defense himself very hard, trying to prove he is just a villager.\n\nHowever, if that person is a werewolf, usually they focusing on how to pretend to be innocent instead of focusing on defending themselves very hard. \n\nSo right now, peoples found some possible werewolves in this competition, but we didn't see any convincing defense from them. That is interesting.\n\n> **n01z3 wrote**\n> \n> > All your comments look very emotional and not very constructive. I do not want to waste time on this kind of discussion.\n> All I can say about public results: seems that I overfitted. I have done this many times and not always successfully. TGS Salt: 11th (in Gold) Public -> 40th Private; Airbus: 2nd (!) Public -> 36th Private.\n> This time, more fortunate. And in the iMaterialist Fashion competition I unintentionally acted without local validation at all and the public LB correlated very well with the private LB.\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/95223#550031\n\n> **ValeriyBabushkin wrote**\n> \n> >Nothing new for me\nIn this competition, I dropped from 3-rd to 129\n\n> >And here from 2 to 128\n> >https://mlbootcamp.ru/round/12/rating/?results_filter=preliminary \n> > https://mlbootcamp.ru/round/12/rating/?results_filter=final\n\n> >129 - 3 = 128 - 2 (coincedence, ins't it ?)\n\n> >On Airbus from 7 to 47\n> >https://www.kaggle.com/c/airbus-ship-detection/leaderboard\n\n> >Here from 1 st to 35\n> >https://mlbootcamp.ru/round/11/rating/?results_filter=preliminary\n> >https://mlbootcamp.ru/round/11/rating/?results_filter=final\n\n> >That is basically the reason why I prefer to wait until the final results are revealed\n\n> >How I do this?\n> >Each time is different. Sometimes it is 5-6 layers stacking, sometimes it is a false correlation with the data, sometimes it is a surplus of computing power in terms of DGX\n\n> >But basically, that I was speaking about the leak, I was thinking about 0.7 + Each time I see someone so high - I am thinking about the leak\n\n> **ValeriyBabushkin wrote**\n> \n>>I would assemble everything into a working solution if I have to provide it, however, I am not in the medals and don't have to\n\n> **ValeriyBabushkin wrote**\n> \n>>don't think place outside medals could be reviewed as strong\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#549863\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#549901\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94430#550001", "550709": "I feel very disappointed about the platform, they release the medals without any explanation of the private sharing things. ", "550700": "All the teams did not drop out places except X5.\n\nAll the teams did not lost points compare with public except X5.\n\nX5 is outliers in this competition\n\nWhat X5 did was an insult to statistics.\n\nIt's hard to imagine that the GM/M in X5 have never learned statistics.", "550665": "Really constructive post :)\n\n> I know one key point in the game is to observe how every player defense themselves.\n\n> In many occasions, ...\n\nAnd I played this game and not every player defends herself like you describe. So what?..) Considering your example, I think you should agree that a lot of clever people understand this logic you suggested and do everything conversely. Why not? So this is very subjective)\n\nAbout this whole situation - not to you personally but for everyone literally exploding (sometimes even with extremely dirty talks from their mouths) who probably will read it - really guys? So if you bomb forums, presenting yourself as very offended people, you really think you are smart or what?) This whole situation looks terribly similar with what happens in the world more and more, when politicians blame the others without any evidence. Now - OK, but how a normal person would buy it then? HOW CAN YOU SERIOUSLY BLAME A PERSON WITHOUT A PROOF WOULD BE FOUND IN A FEW DAYS IF IT EXISTS? I agree that these people blamed at the forum act strange - but these are only GUESSES, not PROOFS. Moving back to our politicians - so either they are dumb, or they are smart and do this to discredit someone. But you all are not politicians, right? So on which side are you, 1 or 2?)) Think about it - if Kaggle officials agreed to take a look at the results, maybe we should wait for the results? If not, then write to them and notify about it) Otherwise your *shoot* at the discussions is pathetic and makes you look miserable, probably you have to drop it in some other well-known place))\n\np.s. meow meow meo meow", "550374": "Thanks for summarizing it. But don’t forget that werewolf board game is based on a russian game called Mafia " } }