Welcome to the first Kaggle competition to detect continuous gravitational waves.
\nWe are scientists from an international research collaboration seeking to push the boundaries of gravitational-wave and geophysics research through the application of machine learning (G2Net). We are also active members of the LIGO-Virgo-KAGRA collaboration, which is responsible for the first detection of gravitational waves from compact binary coalescenses back in 2015. In fact, you may remember some of us from a previous competition focused on that specific kind of signals.
\nIn this case, we need your help to look for continuous gravitational wave signals, which are longer, weaker, and, in summary, harder to find; as it happens, we haven't found any of them! (Yet!) The expected source of this sort of signal is a rapidly-spinning neutron star with some form of imperfection, as explained in this science summary.
\nWe have created a test set comprised of both simulated and actual detector data from LIGO Hanford and LIGO Livingston in which simulated continuous-wave signals were added. For all we know, however, it could be the case that an actual continuous-wave signal is present in the real data and our search methods were not sensitive enough to find it, meaning you could be the first ever to detect a continuous gravitational wave!
\nWe hope you enjoy tackling this challenge and really appreciate your help with such a problem.
\nHave fun!
", "messages": [ { "id": 1969505, "postDate": "2022-10-03T14:24:55.720Z", "content": "Welcome to the first Kaggle competition to detect continuous gravitational waves.
\nWe are scientists from an international research collaboration seeking to push the boundaries of gravitational-wave and geophysics research through the application of machine learning (G2Net). We are also active members of the LIGO-Virgo-KAGRA collaboration, which is responsible for the first detection of gravitational waves from compact binary coalescenses back in 2015. In fact, you may remember some of us from a previous competition focused on that specific kind of signals.
\nIn this case, we need your help to look for continuous gravitational wave signals, which are longer, weaker, and, in summary, harder to find; as it happens, we haven't found any of them! (Yet!) The expected source of this sort of signal is a rapidly-spinning neutron star with some form of imperfection, as explained in this science summary.
\nWe have created a test set comprised of both simulated and actual detector data from LIGO Hanford and LIGO Livingston in which simulated continuous-wave signals were added. For all we know, however, it could be the case that an actual continuous-wave signal is present in the real data and our search methods were not sensitive enough to find it, meaning you could be the first ever to detect a continuous gravitational wave!
\nWe hope you enjoy tackling this challenge and really appreciate your help with such a problem.
\nHave fun!
", "rawMarkdown": "Welcome to the first Kaggle competition to detect continuous gravitational waves.\n\nWe are scientists from an international research collaboration seeking to push the boundaries of gravitational-wave and geophysics research through the application of machine learning [(G2Net)](https://www.g2net.eu/). We are also active members of the LIGO-Virgo-KAGRA collaboration, which is responsible for the first detection of gravitational waves from compact binary coalescenses back in 2015. In fact, you may remember some of us from [a previous competition focused on that specific kind of signals](https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/overview).\n\nIn this case, we need your help to look for *continuous* gravitational wave signals, which are longer, weaker, and, in summary, harder to find; as it happens, we haven't found any of them! (Yet!) The expected source of this sort of signal is a rapidly-spinning neutron star with some form of imperfection, as explained in [this science summary](https://www.ligo.org/science/Publication-O3AllSkyCW/).\n\nWe have created a test set comprised of both simulated and actual detector data from LIGO Hanford and LIGO Livingston in which simulated continuous-wave signals were added. For all we know, however, it could be the case that an *actual* continuous-wave signal is present in the real data and our search methods were not sensitive enough to find it, meaning you could be the first ever to detect a continuous gravitational wave!\n\nWe hope you enjoy tackling this challenge and really appreciate your help with such a problem.\n\nHave fun!\n", "votes": 25 }, { "id": 1972651, "postDate": "2022-10-05T09:05:59.850Z", "content": "Hello @rodrigotenorio there is a typo in the Data tab description : \"target_labels.csv\" should be replaced with \"train_labels.csv\" . Thank you for this interesting competition
", "rawMarkdown": "Hello @rodrigotenorio there is a typo in the Data tab description : \"*target_labels.csv*\" should be replaced with \"*train_labels.csv*\" . Thank you for this interesting competition", "votes": 1, "replies": [ { "id": 1974579, "postDate": "2022-10-06T10:23:12.643Z", "content": "Thanks for spotting this @anisayari, it should be fixed now.
", "rawMarkdown": "Thanks for spotting this @anisayari, it should be fixed now." } ] }, { "id": 2015385, "postDate": "2022-11-03T08:46:54.107Z", "content": "I'm sorry if this is answered elsewhere; the train set synthetic and only the test set is the real data, is this understanding right?
Hello @rodrigotenorio there is a typo in the Data tab description : \"target_labels.csv\" should be replaced with \"train_labels.csv\" . Thank you for this interesting competition
", "votes": 1, "replies": [ { "id": 1974579, "author_name": "Michael J. Williams", "author_url": "", "post_date": "2022-10-06T10:23:12.643000", "content": "Thanks for spotting this @anisayari, it should be fixed now.
", "votes": 0, "replies": [] } ] }, { "id": 2015385, "author_name": "arutema47", "author_url": "", "post_date": "2022-11-03T08:46:54.107000", "content": "I'm sorry if this is answered elsewhere; the train set synthetic and only the test set is the real data, is this understanding right?