deep learning

0. Reference https://arxiv.org/abs/2209.03032 Machine Learning Students Overfit to OverfittingOverfitting and generalization is an important concept in Machine Learning as only models that generalize are interesting for general applications. Yet some students have trouble learning this important concept through lectures and exercises. In this paperarxiv.org1. Introduction- 해당 논문은 Overfitting에 대해..

0. Referencehttps://arxiv.org/abs/1803.08494 Group NormalizationBatch Normalization (BN) is a milestone technique in the development of deep learning, enabling various networks to train. However, normalizing along the batch dimension introduces problems --- BN's error increases rapidly when the batch size becomes smallarxiv.org 1. Introduction- Batch Normalization은 딥러닝에서 흔히 쓰이는 기법중 하나이다.- 하지만 이러..

0. Referencehttps://arxiv.org/abs/1607.06450 Layer NormalizationTraining state-of-the-art, deep neural networks is computationally expensive. One way to reduce the training time is to normalize the activities of the neurons. A recently introduced technique called batch normalization uses the distribution of the summedarxiv.org1. Introduction- DNN은 여러 분야 즉, CV나 NLP문제에 대해 좋은 성과를 보인다.- 하지만, DNN은 학습..

0. Referencehttps://www.sciencedirect.com/science/article/abs/pii/S016786550500303X An introduction to ROC analysisReceiver operating characteristics (ROC) graphs are useful for organizing classifiers and visualizing their performance. ROC graphs are commonly used …www.sciencedirect.com1. Introduction- Receiver Operating Characteristics(ROC) graph는 classifier의 성능을 시각화, 조직화하는 기술이다.- 최근 몇년 동안 Mach..