(Migrated from old blog) Recently, I saw new research which I was amazed by the results and idea behind it. I spent some time trying to read it and hence I am sharing my notes and understanding of...
RecSys Netflix prize competition
(Migrated from old blog) So, I have been taking this course this semester called as Recommendation Systems and the earliest boom this program took was when in 2010 or 2011 Netflix started a compet...
Generative Models - Part II GANs (old)
New Edit (Jan 2023) This post does not talks about other properties of GANs like backward KL divergence, mode collapse, etc. I will write a new post on that soon. (Migrated from old blog) Cool, ...
Generative Models - Part I VAE
(Migrated from old blog) In Generative Models, we generate data based on the distribution of the sample data. basically, we want to learn $ p_{model}(x) $ which is similar to $ p_{data} (x) $ ...
Recurrent neural networks (RNNs)
(Migrated from old blog) One of the abundantly used Neural Networks for Sequential Data or Time Series Data. What do we mean when we say sequential data lets state a few examples of sequential dat...
Keras Callbacks
(Migrated from old blog) Alright, I was on Summer Vacations to my home, India. Therefore, I was inactive for a month or so but now let’s get back and begin again. So что такое ( What is ) Callbac...
Defining model complexity and its math
(Migrated from old blog) I will recommend you first read my post about regularization here : Regularization in Neural Networks So we are done with basics of regularization we know we will add so...
Optimization Algorithms – Gradient Descent Extensions
(Migrated from old blog) We mentioned backpropagation in logistic regression and neural network, In that we used Gradient Descent and a fixed learning rate $\alpha $, for the method to go look for...
Keras LSTM-donald trump
(Migrated from old blog) Okay! Its been long Since I posted something with programming today will be a programming post, lets begin, we will be generating Trump like tweets for Russia because I am...
Kernels 101
(Migrated from old blog) So Kernel is any function $f(x,\bar{x}) $ which represent the scalar product of two variables, It determines the distance in the space. They have two major properties that...