목록딥러닝 (8)

iMTE

Ablation-CAM: Visual Explanations for Deep Convolutional Network Via Gradient-free Localization [XAI-12]

논문 제목 : Ablation-CAM: Visual Explanations for Deep Convolutional Network Via Gradient-free Localization 논문 주소 : https://openaccess.thecvf.com/content_WACV_2020/html/Desai_Ablation-CAM_Visual_Explanations_for_Deep_Convolutional_Network_via_Gradient-free_Localization_WACV_2020_paper.html WACV 2020 Open Access Repository Ablation-CAM: Visual Explanations for Deep Convolutional Network via Gradient-..
Deep learning study/Explainable AI, 설명가능한 AI 2021. 6. 17. 14:32
주요 기계학습, 딥러닝 activation function

activation functions In [1]: import matplotlib.pyplot as plt import numpy as np Step function¶ In [2]: def step(x): return 1*(x>0) In [3]: inputs = np.arange(-5,5,0.01) outputs = step(inputs) plt.figure(figsize=(8,5)) plt.plot(inputs,outputs,label='Step function') plt.hlines(0,-5,5) plt.vlines(0,0,1) plt.xlabel('input',fontsize=24) plt.ylabel('output',fontsize=24) plt.grid(alpha=0.3) plt.title("..
Python/유용한 코드들 2021. 1. 26. 14:42
Meta-learning with Implicit Gradients (NIPS 2019) [1]

Meta-learning with Implicit Gradients [1] https://papers.nips.cc/paper/2019/hash/072b030ba126b2f4b2374f342be9ed44-Abstract.html IntroductionMeta-learning의 frame에서 bi-level optimization procedure는 다음으로 나누어진다.1) inner optimization : 주어진 task에 base learner가 학습하는 과정2) outer optimization : 여러 tasks 들에서 meta learner가 학습하는 과정MAML, DAML, Reptile 등의 방법이 optimization-based methods에 속한다. (Hands-on one-shot..
Deep learning study/Meta-Learning 2020. 12. 24. 15:23
Meta-Learning 2. Optimization as a model for few-shot learning

Paper : OPTIMIZATION AS A MODEL FOR FEW-SHOT LEARNINGDownload : https://openreview.net/forum?id=rJY0-Kcll¬eId=ryq49XyLgAbstract큰 data domain에서 deep neural networks는 큰 성공을 보여주었지만, few-shot learning tasks에서는 성능이 별로 좋지 않았다. (이는 각 class의 매우 적은 example을 보고 빠르게 일반화를 해야하기 때문이다.) 큰 network (e.g., high capacity classifier)에서 gradient-based optimization은 많은 example에 많은 iteration을 해야 성능이 잘나온다는 것이 일반적인 믿..
Deep learning study/Meta-Learning 2019. 4. 20. 19:47
Meta-Learning 1. Learning to learn by gradient descent by gradient descent

Paper : Learning to learn by gradient descent by gradient descentDownload : http://papers.nips.cc/paper/6460-learning-to-learn-by-gradient-descent-by-gradient-descentMeta-Learning을 공부하고 연구에 적용해볼 생각으로 정리해보려고 한다. 꽤 재밌는 아이디어고, 어느 tasks에 적용할 수 있다는 점에서 매우 강력한 알고리즘이 될 것이라고 생각한다. Abstract기존의 hand-crafted features 대신에 학습된 features로의 변화는 꽤 큰 성공이었다. 하지만, 최적화 알고리즘 (e.g., optimization algorithm)은 여전히 hand로 ..
Deep learning study/Meta-Learning 2019. 4. 20. 14:34
Emotion Recognition Using Deep Neural Network and Extreme Learning Machine (2014)

Han, Kun, Dong Yu, and Ivan Tashev. "Speech emotion recognition using deep neural network and extreme learning machine." Fifteenth Annual Conference of the International Speech Communication Association. 2014. Abstract 1. Speech emotion recognition은 어떤 feature들이 의미가 있는지 파악하기 어려움.2. Deep Neural Networks (DNN)은 raw data에서 high-level feature를 추출하여서 speech emotion recognition에 효과적인 결과를 보여줌.3. Extrem..
Deep learning 2018. 5. 28. 17:04
Class Imbalance Problem in Data Mining : Review

Longadge, Rushi, and Snehalata Dongre. "Class imbalance problem in data mining review." arXiv preprint arXiv:1305.1707(2013). https://arxiv.org/ftp/arxiv/papers/1305/1305.1707.pdf Classification of data becomes difficult because of unbounded size and imbalance nature of data. Class imbalance problem become greatest issue in data mining. Data의 imbalance함은 두개의 class 중 하나의 class를 더 많이 sampling해서 문제..
Machine learning 2018. 2. 27. 22:28
Deep learning 공부 유용 자료들

김성훈 교수님의 모두의 딥러닝 https://hunkim.github.io/ml/ 우재준님의 blog http://jaejunyoo.blogspot.com/2017/04/began-boundary-equilibrium-gan-1.html 이기창님의 blog https://ratsgo.github.io/machine%20learning/2017/05/23/SVM/ 최성준님의 edwith 강의 http://www.edwith.org/deeplearningchoi Standford 수업 CS231n http://cs231n.stanford.edu/ Udacity deep learning 무료 강의 https://www.udacity.com/course/deep-learning--ud730 김태영님의 keras..
Deep learning 2018. 2. 11. 12:30
Prev 1 Next