Machine Learning for InfoSec

Transcript

1. Machine Learning for InfoSec
2. content • • • • • • Logistic • • • • •
3. Why Python ? • • • • • •
4. Machine Learning
6. Wanna Start ML ...
7. What is Machine Learning ? • •
8. Supervised learning Unsupervised learning Map Input variables/features to some discrete value Map Input variables/features to some Continuous value Data analysis to find hidden patterns/grouping in data
10. Linear Regression
11. Learning (determining) good values for all the weights and the bias from labeled examples (With minimum loss) that model can able to predict new unseen data value. Given an X (input Features) and Y (target/label) GOAL
12. Linear regression - Theory
13. Loss • loss is a number indicating how bad the model's prediction was on a single example. If the model's prediction is perfect, the loss is zero; otherwise, the loss is greater.
15. Linear Regression with One Variable Problem : Here we will implement linear regression with one variable to predict profits for a food truck. Suppose you are the CEO of a restaurant franchise and are considering different cities for opening a new outlet. The chain already has trucks in various cities and you have data for profits and populations from the cities. http://bit.ly/linearReg (colab) File-> Save a copy in drive http://bit.ly/datasetex1(Dataset)
17. No, I don’t like loss let’s reduce it. But how ??
18. Gradient descent
20. Return to colab http://bit.ly/linearReg Another approach
22. Multiple Linear Regression(Multiple Features)
23. Logistic Regression
24. Theory Instead of predicting exactly 0 or 1, logistic regression generates a probability—a value between 0 and 1, exclusive Many problems require a probability estimate as output. Logistic regression is an extremely efficient mechanism for calculating probabilities. Log Loss is the loss function for logistic regression.
25. Sigmoid function
26. Logistic Regression using Python (scikit-learn) http://bit.ly/log987
27. Neural Network
28. Perceptron Perceptron is a single layer neural network and a multi-layer perceptron is called Neural Networks. Perceptron is a linear classifier (binary). Also, it is used in supervised learning.
29. A neural network is nothing more than a bunch of neurons connected together.
30.   
31. Types of Neural Network • • • •
32. Let’s try to understand by Visualization
33. Build basic ANN from Scratch
34. Machine Learning for InfoSec
35. InfoSec Information Security • • •
36. Real-time use cases • •
37. Top Use Cases • • • • •
38. There will always be a man trying to find weaknesses in systems or ML algorithms and to bypass security mechanisms. What’s worse, now hackers are able to use machine learning to carry out all their nefarious endeavors.[1]
39. Machine Learning for Email and SMS Spam detection Note: Create a copy in drive
41. Machine Learning for Malware Analysis
42. All Links and Material https://goo.gl/3L9cGu