Have you ever thought about how certain you are about the occurring events or decisions you make. If you have, let’s talk more detailed.

Firstly, let’s start with “what is uncertainty?” . “Uncertainty refers to epistemic situations involving imperfect or unknown information. It applies to predictions of future events, to physical measurements that are already made, or to the unknown”. That is the definition from Wikipedia.

Basically it is the measure of how much you sure about something. For example , if you know that there are 3 red balls in the box, and if someone ask you to pick…

Resampling methods are very useful and beneficial in statistics and machine learning to fit more accurate models, model selection and parameter tuning.They draw samples from train data and fit model to check the variability of model and get additional information.We cannot best sure of the result of the model by just unique fit without testing on different sample or samples. It can be computationally expensive because of fitting model more than one, but recent improvements tackle this issue easily without too much effort.

In this post I will talk about one of the resampling method which is called cross-validation and…

Gradient Descent is iterative optimization algorithm , which provides new point in each iteration based on its gradient and learning rate that we initialise at the beginning.

Gradient is the vector of the objective function that contains the partial derivatives of function respect to the coordinates or dimensions of the point.

Learning rate identify our step size ,that means in which amount we want to get close to the optimal point. The direction of the step is the negative of the gradient in minimization problems, and positive of gradient in the maximisation problems.There …

Industrial Engineer with focus on Machine Learning, Optimization and Deep Learning. I love to share my knowledge and ideas :)