본문 바로가기

AI20

[KDD-2019] Time-Series Anomaly Detection Service at Microsoft 논문 전문 : https://arxiv.org/abs/1906.03821 [출처] KDD '19: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data MiningJuly 2019Pages 3009–3017https://doi.org/10.1145/3292500.3330680 ※ The picture and content of this article are from the original paper. This article is more of an intuitive understanding than academic analysis. [논문 요약] Time-Series Anomaly Detection.. 2024. 1. 22.
[AAAI-2019] TabNet : Attentive Interpetable Tabular Learning 논문 전문 : https://arxiv.org/abs/1908.07442 [출처] https://doi.org/10.1609/aaai.v35i8.16826 ※ The picture and content of this article are from the original paper. This article is more of an intuitive understanding than academic analysis. [논문 요약] TabNet : Attentive Interpretable Tabular Learning Tabular Data를 다루는 분들에게는 꽤나 유명한(Citation 800이상) 구글에서 나온 TabNet 논문입니다. 당시에는 Tab Data 기준으로는 SOTA 성능이였습니다만, Tab.. 2024. 1. 18.
[ICML-2021] Tabular Data:Deep Learning is Not All You Need 논문 전문 : https://arxiv.org/abs/2106.03253 [출처] Ravid Shwartz-Ziv, Amitai Armon,Tabular data: Deep learning is not all you need,Information Fusion,Volume 81,2022,Pages 84-90,ISSN 1566-2535,https://doi.org/10.1016/j.inffus.2021.11.011. ※ The picture and content of this article are from the original paper. This article is more of an intuitive understanding than academic analysis. [논문 요약] Tabular Dat.. 2024. 1. 11.
[NIPS-2017] Attention is all you need 논문 전문 : https://arxiv.org/abs/1706.03762 [출처] https://doi.org/10.48550/arXiv.1706.03762 ※ The picture and content of this article are from the original paper. This article is more of an intuitive understanding than academic analysis. [논문 정리] Attention is all you need Attention is all you need라는 제목만으로도 오금이 저릴정도로 슈퍼 논문입니다. 제가 생각하기에는, GAN과 더불어 2000년대 나온 최고의 논문이 아닐까 합니다. 후대에는 이 시점을 AGI(Artificial Ge.. 2023. 12. 23.
[CVPR-2018] ESRGAN : Enhanced Super-Resolution Generative Adversarial Networks 논문 전문 : https://arxiv.org/abs/1809.00219 [출처] https://doi.org/10.48550/arXiv.1809.00219 ※ The picture and content of this article are from the original paper. This article is more of an intuitive understanding than academic analysis. [논문 요약] ESRGAN : Enhanced Super-Resolution Generative Adversarial Networks 본 논문은 GAN을 활용한 Super Resolution(이하 SR)에 거의 시초격인 논문입니다. GAN 네트워크야 CNN으로 짜면 얼마든지 원하는 모델을 만들.. 2023. 12. 11.