Deep learning for Junior Developer
0 likes92 views
This document provides an overview of deep learning for beginners. It discusses how deep learning has evolved from rule-based systems to neural networks. It then explains the basic mathematical concepts in machine learning including linear regression, hypothesis, cost/loss functions, and gradient descent. Gradient descent is used to minimize the cost function and train the model by iteratively updating the weights and biases. The trained model can then be tested on new input data to check if it can accurately predict the expected output.
![X = [1, 2, 3]
Y = [1, 2, 3]
X = 4 일때 Y = ?](https://image.slidesharecdn.com/deeplearning-180502053721/85/Deep-learning-for-Junior-Developer-10-320.jpg)
![Hypothesis
H(X) = X*W + B
X = [1, 2, 3] Y = [1, 2, 3]
W=2 , B = 0
H(X) = [H(1), H(2), H(3)] = [2, 4, 6]](https://image.slidesharecdn.com/deeplearning-180502053721/85/Deep-learning-for-Junior-Developer-14-320.jpg)
![Cost / Loss function
H(X) = [H(1), H(2), H(3)] = [2, 4, 6]
Y = [1, 2, 3]
euclidean distance](https://image.slidesharecdn.com/deeplearning-180502053721/85/Deep-learning-for-Junior-Developer-15-320.jpg)
![Cost / Loss function
X = [1, 2, 3] Y = [1, 2, 3]
W= 2, B=0 H = [2, 4, 6] cost(2) = 14
W= 1.5, B=0 H = [1.5, 3, 4.5] cost(1.5) = 1.166
W= 1, B=0 H = [1, 2, 3] cost(1) = 0](https://image.slidesharecdn.com/deeplearning-180502053721/85/Deep-learning-for-Junior-Developer-16-320.jpg)
![Cost / Loss function
X = [1, 2, 3] Y = [1, 2, 3] B = 0](https://image.slidesharecdn.com/deeplearning-180502053721/85/Deep-learning-for-Junior-Developer-17-320.jpg)
![Gradient Descent
W= 2, B=0 H = [2, 4, 6] cost(2) = 14
W= 1.5, B=0 H = [1.5, 3, 4.5] cost(1.5) = 1.166
W= 1, B=0 H = [1, 2, 3] cost(1) = 0
Minimize Cost function](https://image.slidesharecdn.com/deeplearning-180502053721/85/Deep-learning-for-Junior-Developer-18-320.jpg)
![Train & Test
X = [1, 2, 3] Y = [1, 2, 3]
Hypothesis
Cost/Loss
Gradient Descent
X = 4 일때 Y = 4 가 맞음?
W, B](https://image.slidesharecdn.com/deeplearning-180502053721/85/Deep-learning-for-Junior-Developer-22-320.jpg)
Deep learning for Junior Developer
- 1. Deep Learning 초보 개발자를 위한 개요
- 4. Internet
- 5. Mobile
- 6. Next?
- 8. Spam Filter • Rule based -> Explicit • Train • Neural Network • Deep, Wide
- 10. X = [1, 2, 3] Y = [1, 2, 3] X = 4 일때 Y = ?
- 11. THE MATHEMATICAL WAY OF THINKING Science 15 Nov 1940:
- 12. Y = X * W + B W: weight, B: bias
- 13. Machine Learning • Hypothesis • cost / loss • Gradient descent • Train • Test
- 14. Hypothesis H(X) = X*W + B X = [1, 2, 3] Y = [1, 2, 3] W=2 , B = 0 H(X) = [H(1), H(2), H(3)] = [2, 4, 6]
- 15. Cost / Loss function H(X) = [H(1), H(2), H(3)] = [2, 4, 6] Y = [1, 2, 3] euclidean distance
- 16. Cost / Loss function X = [1, 2, 3] Y = [1, 2, 3] W= 2, B=0 H = [2, 4, 6] cost(2) = 14 W= 1.5, B=0 H = [1.5, 3, 4.5] cost(1.5) = 1.166 W= 1, B=0 H = [1, 2, 3] cost(1) = 0
- 17. Cost / Loss function X = [1, 2, 3] Y = [1, 2, 3] B = 0
- 18. Gradient Descent W= 2, B=0 H = [2, 4, 6] cost(2) = 14 W= 1.5, B=0 H = [1.5, 3, 4.5] cost(1.5) = 1.166 W= 1, B=0 H = [1, 2, 3] cost(1) = 0 Minimize Cost function
- 20. Gradient Descent
- 22. Train & Test X = [1, 2, 3] Y = [1, 2, 3] Hypothesis Cost/Loss Gradient Descent X = 4 일때 Y = 4 가 맞음? W, B
- 23. Thank you