Hello !

Maxout - Learning Activation Function

You must have heard of different activation functions in our deep learning models, starting from sigmoid function to ReLu, LeakyReLU, etc. But how do we know that which is the best activation function? Well, the Maxout unit is an activation function which is itself trained by our model.

L2 Regularization: The mathematics behind It

L2 regularisation, one of the most popular techniques in Machine Learning, is a method to reduce the variance of the model and increase the bias to make the model more generalisable. L2 regularisation is a MAP estimation with Gaussian prior probabilities.

Why "early-stopping" works as Regularization?

While training our machine learning models, we often observe that with a model with significant learning capacity training error steadily decreases but after some epochs validation error starts increasing. After each epoch, the model learns the data and updates the weights accordingly. Training and validation error decreases as long as our model is generalising the input data.

Using DS1302 RTC module with Arduino

The Arduino can keep track of time very accurately using the millis() function, but only since the moment, it was powered up. If the Arduino is powered off or a reset is made, the Arduino timer starts with 0. This feature of Arduino is not very useful in case real-time is required in Arduino projects. We solve this problem using a Real-Time Clock (RTC) module.

The Mathematics Behind Luck

Quite often we feel disappointed with the results of the tasks we have done sheer hard work on. The results just don’t turn up as expected in proportion to the amount of hard work we put in and directly hit our motivation for our upcoming tasks. What usually follows is a period of self-guilt extended over hours, over days,and sometimes over weeks as well. How can I be so stupid?