Archive - June 17, 2019

Table Of content Simple Explanation of Adaboost Step by step understanding of how adaboost works. Number of weak learners required Bias and variance tradeoff in adaboost Parameter optimization in adaboost Feature selection in adaboost SIMPLE EXPLANATION OF ADABOOST Adaboost creates an ensemble of weak learners to create a strong learner. Weak learners are models that achieve accuracy just above random chance on a classification problem. The most suited and therefore most common algorithm used with AdaBoost are decision trees with one level. Because these trees are so short and only contain one decision for classification, they are often called decision stumps. Adboost is a sequential learner. Basically, in adaboost you run all of your data through a weak learner, and try to classify the data. Then in the next iteration, you give more weightage to the incorrectly classified examples in the training data. So, your next weak learner does a[…]

Table of content Understanding how you calculate distance to a point from a plane Maths behind how you find support vectors How does and why does a kernel work How to figure out if your support vector is good at generalizing Bias and variance tradeoff in support vector machines Imbalanced Classes In SVM Loss function in svm Assumptions under SVM and stuff to be careful about SIMPLEST WAY TO IMAGINE SUPPORT VECTOR MACHINES Take this plotted data for a 2 dimensional data with two different labels. Now try to imagine a line somewhere in the middle that differentiates both the data sets. I am pretty sure most of you can picture this line. Our aim with SVM is to find this exact line. In more technical terms, we want a line or hyper plane so that the distance of the closest points in each group from the line/Hyperplane is the[…]