{ "id": 451732, "title": "Predict AI Model Runtime: Research Papers ", "url": "/competitions/predict-ai-model-runtime/discussion/451732", "author_name": "", "post_date": "2023-10-30T09:39:43.180349800Z", "votes": 6, "comment_count": 1, "views": 0, "content": "

WIth respect to \"Google Fast or slow ? Predict AI Model Runtime\" Competition. I am listing here highly
\nrelevant research papers:

\n
    \n

  1. \"Runtime Performance Prediction for Deep Learning Models with Graph Neural Network\"
    \nYanjie Gao et al. link: https://www.microsoft.com/en-us/research/uploads/prod/2021/02/dnnperf.pdf

    2. \n

  2. Frank Hutter, Lin Xu, Holger H. Hoos, Kevin Leyton-Brown,
    \nAlgorithm runtime prediction: Methods & evaluation, Artificial Intelligence, Volume 206, 2014,
    \nPages 79-111, ISSN 0004-3702, https://doi.org/10.1016/j.artint.2013.10.003.
    \nlink : https://www.sciencedirect.com/science/article/pii/S0004370213001082

    3. \n

  3. Xinyi Li et al, \"Use Machine Learning to Predict the Running Time of the Program\"
    \nCSAI '21: Proceedings of the 2021 5th International Conference on Computer Science and Artificial Intelligence December 2021 Pages 260–266 https://doi.org/10.1145/3507548.3507588
    \nLink https://dl.acm.org/doi/abs/10.1145/3507548.3507588

    4. \n
\n

Appreciate sharing of information on useful research articles for the competition problem statement.

", "messages": [ { "id": "2504977", "postDate": "10/30/2023 09:39:43", "content": "

WIth respect to \"Google Fast or slow ? Predict AI Model Runtime\" Competition. I am listing here highly
\nrelevant research papers:

\n
    \n

  1. \"Runtime Performance Prediction for Deep Learning Models with Graph Neural Network\"
    \nYanjie Gao et al. link: https://www.microsoft.com/en-us/research/uploads/prod/2021/02/dnnperf.pdf

    2. \n

  2. Frank Hutter, Lin Xu, Holger H. Hoos, Kevin Leyton-Brown,
    \nAlgorithm runtime prediction: Methods & evaluation, Artificial Intelligence, Volume 206, 2014,
    \nPages 79-111, ISSN 0004-3702, https://doi.org/10.1016/j.artint.2013.10.003.
    \nlink : https://www.sciencedirect.com/science/article/pii/S0004370213001082

    3. \n

  3. Xinyi Li et al, \"Use Machine Learning to Predict the Running Time of the Program\"
    \nCSAI '21: Proceedings of the 2021 5th International Conference on Computer Science and Artificial Intelligence December 2021 Pages 260–266 https://doi.org/10.1145/3507548.3507588
    \nLink https://dl.acm.org/doi/abs/10.1145/3507548.3507588

    4. \n
\n

Appreciate sharing of information on useful research articles for the competition problem statement.

", "rawMarkdown": "WIth respect to \"Google Fast or slow ? Predict AI Model Runtime\" Competition. I am listing here highly\nrelevant research papers:\n1. \"Runtime Performance Prediction for Deep Learning Models with Graph Neural Network\"\nYanjie Gao et al. link: [https://www.microsoft.com/en-us/research/uploads/prod/2021/02/dnnperf.pdf](url)\n\n2. Frank Hutter, Lin Xu, Holger H. Hoos, Kevin Leyton-Brown,\nAlgorithm runtime prediction: Methods & evaluation, Artificial Intelligence, Volume 206, 2014,\nPages 79-111, ISSN 0004-3702, https://doi.org/10.1016/j.artint.2013.10.003.\nlink : [https://www.sciencedirect.com/science/article/pii/S0004370213001082](url)\n\n3. Xinyi Li et al, \"Use Machine Learning to Predict the Running Time of the Program\"\nCSAI '21: Proceedings of the 2021 5th International Conference on Computer Science and Artificial Intelligence December 2021 Pages 260–266 https://doi.org/10.1145/3507548.3507588\nLink [https://dl.acm.org/doi/abs/10.1145/3507548.3507588](url)\n\nAppreciate sharing of information on useful research articles for the competition problem statement.", "votes": null }, { "id": "2510591", "postDate": "11/03/2023 05:51:47", "content": "

Also, sharing the relevant paper on TpuGraphs Data set which is valuable for this competition:
\n\"TpuGraphs: A Performance Prediction Dataset on Large Tensor Computational Graphs\" (arXiv:2308.13490)
\nhttps://arxiv.org/pdf/2308.13490.pdf

", "rawMarkdown": "Also, sharing the relevant paper on TpuGraphs Data set\t which is valuable for this competition:\n\"TpuGraphs: A Performance Prediction Dataset on Large Tensor Computational Graphs\" (arXiv:2308.13490) \n[https://arxiv.org/pdf/2308.13490.pdf](url)", "votes": null } ], "comments": [ { "id": 2510591, "author_name": "crsuthikshnkumar", "author_url": "", "post_date": "11/03/2023 05:51:47", "content": "

Also, sharing the relevant paper on TpuGraphs Data set which is valuable for this competition:
\n\"TpuGraphs: A Performance Prediction Dataset on Large Tensor Computational Graphs\" (arXiv:2308.13490)
\nhttps://arxiv.org/pdf/2308.13490.pdf

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2504977": "WIth respect to \"Google Fast or slow ? Predict AI Model Runtime\" Competition. I am listing here highly\nrelevant research papers:\n1. \"Runtime Performance Prediction for Deep Learning Models with Graph Neural Network\"\nYanjie Gao et al. link: [https://www.microsoft.com/en-us/research/uploads/prod/2021/02/dnnperf.pdf](url)\n\n2. Frank Hutter, Lin Xu, Holger H. Hoos, Kevin Leyton-Brown,\nAlgorithm runtime prediction: Methods & evaluation, Artificial Intelligence, Volume 206, 2014,\nPages 79-111, ISSN 0004-3702, https://doi.org/10.1016/j.artint.2013.10.003.\nlink : [https://www.sciencedirect.com/science/article/pii/S0004370213001082](url)\n\n3. Xinyi Li et al, \"Use Machine Learning to Predict the Running Time of the Program\"\nCSAI '21: Proceedings of the 2021 5th International Conference on Computer Science and Artificial Intelligence December 2021 Pages 260–266 https://doi.org/10.1145/3507548.3507588\nLink [https://dl.acm.org/doi/abs/10.1145/3507548.3507588](url)\n\nAppreciate sharing of information on useful research articles for the competition problem statement.", "2510591": "Also, sharing the relevant paper on TpuGraphs Data set\t which is valuable for this competition:\n\"TpuGraphs: A Performance Prediction Dataset on Large Tensor Computational Graphs\" (arXiv:2308.13490) \n[https://arxiv.org/pdf/2308.13490.pdf](url)" }, "source": "meta" }