The competition advises the participants to contribute to summarising the recent advancements made by the community in 7 fields : Text Data, Image/Video Data, Time Series Data, Kaggle Competitions, Generative AI and AI Ethics plus a miscellaneous 'Other' section.
\nSince the essays are meant to showcase a broader review/perspective of advancements in a specified field, the content of the essays will inevitably end up with an overlap between several fields. Events in one field of AI have impacted progress in other fields - for example how Generative AI has sparked an entirely new AI Ethics debate (as well as a worldwide pro/anti generative model campaign) which in turn led to projects like Glaze and cropped up more community interest in latent space inversion and a recent competition by stable diffusion to generate prompts from images.
\nAlthough the extent of the overlap may vary, what would the criteria for quantifying the essays under a specific field be, especially if the focus of the essay touches upon various problems? How will points/prizes be awarded if an essay that does not revolve around a central crux field but touches on various fields since the final winners are to be chosen as being one from each topic - or would it fall under 'Others'? If it's the latter, it may result in a huge stockpile of submissions in the 'Others' section.
This also leads to another question: \"How niche is the essay going to be?\" or rather \"How is topic specificity going to impact grading and popularity?\". Directions provided by the Kaggle team advise us to look at things through a broader perspective which may capture the overall happenings in a field or subfield through a giant lens. However, great essays that are written with the objective to help a layman navigate the depths of a very specific topic may fall behind if attention is drawn towards essays written to cater for a broader perspective. That said, the competition evaluation rules specify other metrics that may play a part in grading aside the choice of topic.
\n[P.S.: Although my account is two years old - this is my first ever proper interaction on Kaggle. Hoping to learn a lot here :D]
", "messages": [ { "id": "2257425", "postDate": "05/13/2023 10:38:47", "content": "The competition advises the participants to contribute to summarising the recent advancements made by the community in 7 fields : Text Data, Image/Video Data, Time Series Data, Kaggle Competitions, Generative AI and AI Ethics plus a miscellaneous 'Other' section.
\nSince the essays are meant to showcase a broader review/perspective of advancements in a specified field, the content of the essays will inevitably end up with an overlap between several fields. Events in one field of AI have impacted progress in other fields - for example how Generative AI has sparked an entirely new AI Ethics debate (as well as a worldwide pro/anti generative model campaign) which in turn led to projects like Glaze and cropped up more community interest in latent space inversion and a recent competition by stable diffusion to generate prompts from images.
\nAlthough the extent of the overlap may vary, what would the criteria for quantifying the essays under a specific field be, especially if the focus of the essay touches upon various problems? How will points/prizes be awarded if an essay that does not revolve around a central crux field but touches on various fields since the final winners are to be chosen as being one from each topic - or would it fall under 'Others'? If it's the latter, it may result in a huge stockpile of submissions in the 'Others' section.
This also leads to another question: \"How niche is the essay going to be?\" or rather \"How is topic specificity going to impact grading and popularity?\". Directions provided by the Kaggle team advise us to look at things through a broader perspective which may capture the overall happenings in a field or subfield through a giant lens. However, great essays that are written with the objective to help a layman navigate the depths of a very specific topic may fall behind if attention is drawn towards essays written to cater for a broader perspective. That said, the competition evaluation rules specify other metrics that may play a part in grading aside the choice of topic.
\n[P.S.: Although my account is two years old - this is my first ever proper interaction on Kaggle. Hoping to learn a lot here :D]
", "rawMarkdown": "The competition advises the participants to contribute to summarising the recent advancements made by the community in 7 fields : *Text Data, Image/Video Data, Time Series Data, Kaggle Competitions, Generative AI and AI Ethics* plus a miscellaneous *'Other'* section. \n\nSince the essays are meant to showcase a broader review/perspective of advancements in a specified field, the content of the essays will *inevitably end up with an overlap between several fields*. Events in one field of AI have impacted progress in other fields - for example how Generative AI has sparked an entirely new AI Ethics debate (as well as a worldwide pro/anti generative model campaign) which in turn led to projects like [Glaze](https://glaze.cs.uchicago.edu/) and cropped up more community interest in latent space inversion and a [recent competition by stable diffusion](https://www.kaggle.com/competitions/stable-diffusion-image-to-prompts) to generate prompts from images. \n\nAlthough the extent of the overlap may vary, what would the criteria for quantifying the essays under a specific field be, especially if the focus of the essay touches upon various problems? How will points/prizes be awarded if an essay that does not revolve around a central crux field but touches on various fields since the final winners are to be chosen as being one from each topic - or would it fall under *'Others'*? If it's the latter, it may result in a huge stockpile of submissions in the *'Others'* section.\n\n# On the Nicheness of an Essay\nThis also leads to another question: *\"How niche is the essay going to be?\"* or rather *\"How is topic specificity going to impact grading and popularity?\"*. Directions provided by the Kaggle team advise us to look at things through a broader perspective which may capture the overall happenings in a field or subfield through a giant lens. However, great essays that are written with the objective to help a layman navigate the depths of a very specific topic may fall behind if attention is drawn towards essays written to cater for a broader perspective. That said, the [competition evaluation rules](https://www.kaggle.com/competitions/2023-kaggle-ai-report/overview/evaluation) specify other metrics that may play a part in grading aside the choice of topic.\n\n### Good luck, everyone.\n\n[P.S.: Although my account is two years old - this is my first ever proper interaction on Kaggle. Hoping to learn a lot here :D]", "votes": null }, { "id": "2260292", "postDate": "05/15/2023 15:29:56", "content": "Q1: Which topic is most appropriate?
\nYou will select for yourself whichever category that you think is most relevant for your essay. The \"other notable advances in ML\" section is meant to review anything that does not fit into any of the other categories, and, field-specific essays that discuss how advancements within multiple areas have impacted a single field is one example that might fit nicely into the \"other\" category. Ultimately you have a lot of autonomy and flexibility to make this determination yourself.
\nQ2: Broad essays vs narrow essays?
\nAn essay that comprehensively summarizes the recent advancements in a given area will likely score higher than an essay that has a comparatively narrow focus. Ultimately the goal is to respond to one of the 7x prompts in a way that will score highest against the evaluation rubric, and the prompt begins with \"what the ML community has learned over the past 2 years of working with…\", which is a fairly broad mandate.
", "rawMarkdown": "Q1: Which topic is most appropriate?\n\nYou will select for yourself whichever category that you think is most relevant for your essay. The \"other notable advances in ML\" section is meant to review anything that does not fit into any of the other categories, and, field-specific essays that discuss how advancements within multiple areas have impacted a single field is one example that might fit nicely into the \"other\" category. Ultimately you have a lot of autonomy and flexibility to make this determination yourself.\n\nQ2: Broad essays vs narrow essays?\n\nAn essay that comprehensively summarizes the recent advancements in a given area will likely score higher than an essay that has a comparatively narrow focus. Ultimately the goal is to respond to one of the [7x prompts](https://www.kaggle.com/competitions/2023-kaggle-ai-report/overview/description) in a way that will score highest against the [evaluation rubric](https://www.kaggle.com/competitions/2023-kaggle-ai-report/overview/evaluation), and the prompt begins with \"what the ML community has learned over the past 2 years of working with...\", which is a fairly broad mandate.", "votes": null }, { "id": "2260377", "postDate": "05/15/2023 16:18:04", "content": "Thank you for your comprehensive reply to my questions! Seems like the free reign given to the competitors is not only a way to ensure the production of high-grade essays but also allows kagglers from diverse backgrounds to contribute to a compiled knowledge repository - this entire process is almost akin to a large social experiment. Whichever way this competition goes, I know we're going to learn a lot from the results and the journey. And again, thanks for looking out for newbies like me @paultimothymooney :D Really appreciate it!
", "rawMarkdown": "Thank you for your comprehensive reply to my questions! Seems like the free reign given to the competitors is not only a way to ensure the production of high-grade essays but also allows kagglers from diverse backgrounds to contribute to a compiled knowledge repository - this entire process is almost akin to a large social experiment. Whichever way this competition goes, I know we're going to learn a lot from the results and the journey. And again, thanks for looking out for newbies like me @paultimothymooney :D Really appreciate it!", "votes": null } ], "comments": [ { "id": 2260292, "author_name": "paultimothymooney", "author_url": "", "post_date": "05/15/2023 15:29:56", "content": "Q1: Which topic is most appropriate?
\nYou will select for yourself whichever category that you think is most relevant for your essay. The \"other notable advances in ML\" section is meant to review anything that does not fit into any of the other categories, and, field-specific essays that discuss how advancements within multiple areas have impacted a single field is one example that might fit nicely into the \"other\" category. Ultimately you have a lot of autonomy and flexibility to make this determination yourself.
\nQ2: Broad essays vs narrow essays?
\nAn essay that comprehensively summarizes the recent advancements in a given area will likely score higher than an essay that has a comparatively narrow focus. Ultimately the goal is to respond to one of the 7x prompts in a way that will score highest against the evaluation rubric, and the prompt begins with \"what the ML community has learned over the past 2 years of working with…\", which is a fairly broad mandate.
", "votes": null, "replies": [ { "id": 2260377, "author_name": "snowclipsed", "author_url": "", "post_date": "05/15/2023 16:18:04", "content": "Thank you for your comprehensive reply to my questions! Seems like the free reign given to the competitors is not only a way to ensure the production of high-grade essays but also allows kagglers from diverse backgrounds to contribute to a compiled knowledge repository - this entire process is almost akin to a large social experiment. Whichever way this competition goes, I know we're going to learn a lot from the results and the journey. And again, thanks for looking out for newbies like me @paultimothymooney :D Really appreciate it!
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "2257425": "The competition advises the participants to contribute to summarising the recent advancements made by the community in 7 fields : *Text Data, Image/Video Data, Time Series Data, Kaggle Competitions, Generative AI and AI Ethics* plus a miscellaneous *'Other'* section. \n\nSince the essays are meant to showcase a broader review/perspective of advancements in a specified field, the content of the essays will *inevitably end up with an overlap between several fields*. Events in one field of AI have impacted progress in other fields - for example how Generative AI has sparked an entirely new AI Ethics debate (as well as a worldwide pro/anti generative model campaign) which in turn led to projects like [Glaze](https://glaze.cs.uchicago.edu/) and cropped up more community interest in latent space inversion and a [recent competition by stable diffusion](https://www.kaggle.com/competitions/stable-diffusion-image-to-prompts) to generate prompts from images. \n\nAlthough the extent of the overlap may vary, what would the criteria for quantifying the essays under a specific field be, especially if the focus of the essay touches upon various problems? How will points/prizes be awarded if an essay that does not revolve around a central crux field but touches on various fields since the final winners are to be chosen as being one from each topic - or would it fall under *'Others'*? If it's the latter, it may result in a huge stockpile of submissions in the *'Others'* section.\n\n# On the Nicheness of an Essay\nThis also leads to another question: *\"How niche is the essay going to be?\"* or rather *\"How is topic specificity going to impact grading and popularity?\"*. Directions provided by the Kaggle team advise us to look at things through a broader perspective which may capture the overall happenings in a field or subfield through a giant lens. However, great essays that are written with the objective to help a layman navigate the depths of a very specific topic may fall behind if attention is drawn towards essays written to cater for a broader perspective. That said, the [competition evaluation rules](https://www.kaggle.com/competitions/2023-kaggle-ai-report/overview/evaluation) specify other metrics that may play a part in grading aside the choice of topic.\n\n### Good luck, everyone.\n\n[P.S.: Although my account is two years old - this is my first ever proper interaction on Kaggle. Hoping to learn a lot here :D]", "2260292": "Q1: Which topic is most appropriate?\n\nYou will select for yourself whichever category that you think is most relevant for your essay. The \"other notable advances in ML\" section is meant to review anything that does not fit into any of the other categories, and, field-specific essays that discuss how advancements within multiple areas have impacted a single field is one example that might fit nicely into the \"other\" category. Ultimately you have a lot of autonomy and flexibility to make this determination yourself.\n\nQ2: Broad essays vs narrow essays?\n\nAn essay that comprehensively summarizes the recent advancements in a given area will likely score higher than an essay that has a comparatively narrow focus. Ultimately the goal is to respond to one of the [7x prompts](https://www.kaggle.com/competitions/2023-kaggle-ai-report/overview/description) in a way that will score highest against the [evaluation rubric](https://www.kaggle.com/competitions/2023-kaggle-ai-report/overview/evaluation), and the prompt begins with \"what the ML community has learned over the past 2 years of working with...\", which is a fairly broad mandate.", "2260377": "Thank you for your comprehensive reply to my questions! Seems like the free reign given to the competitors is not only a way to ensure the production of high-grade essays but also allows kagglers from diverse backgrounds to contribute to a compiled knowledge repository - this entire process is almost akin to a large social experiment. Whichever way this competition goes, I know we're going to learn a lot from the results and the journey. And again, thanks for looking out for newbies like me @paultimothymooney :D Really appreciate it!" }, "source": "meta" }