일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | ||||||
2 | 3 | 4 | 5 | 6 | 7 | 8 |
9 | 10 | 11 | 12 | 13 | 14 | 15 |
16 | 17 | 18 | 19 | 20 | 21 | 22 |
23 | 24 | 25 | 26 | 27 | 28 |
- AI
- Score-CAM
- GAN
- Machine Learning
- coding test
- 기계학습
- Explainable AI
- 코딩 테스트
- Deep learning
- Cam
- 백준
- Interpretability
- Unsupervised learning
- 머신러닝
- Class activation map
- python
- grad-cam
- 인공지능
- 코딩테스트
- 설명가능한 인공지능
- xai
- 시계열 분석
- meta-learning
- 딥러닝
- cs231n
- SmoothGrad
- Artificial Intelligence
- 메타러닝
- 설명가능한
- keras
- Today
- Total
목록Deep learning/Tensorflow (2)
iMTE
Restricted Boltzmann Machine back propagation을 사용해, deep한 neural network를 학습시킬 수 있다. 하지만, deep networks의 cost function은 non-convex function이므로, 어디서 최적화를 시작하는지는 모델의 성능을 크게 좌우한다. (Weights initialization) 다양한 initializer들이 제시되었다. Xavier initializer, He initializer 등이 그렇다. 이 initializer는 들어오는 neurons의 수와 (fan-in) 출력되는 neurons의 수 (fan-out)을 사용해서 normal distribution의 variance를 결정한다. 이 방법은 대부분의 neuron들..
Sparse regularization Autoencoder는 입력된 데이터를 저차원으로 encoding하고 다시 고차원으로 decoding하는 unsupervised learning이다. 만약 hidden layer가 input layer의 dimension보다 작고, activation function으로 linear function을 사용하면, 이 autoencoder는 PCA와 동일한 역할을 한다. 즉, 분산이 큰 factor를 찾아내고 그 factor로 데이터를 transformation하는 것이다. 만약, hidden layer와 input layer의 차원이 같고 activation function이 linear function이라면, hidden layer는 identity matrix와 ..