Oct 20, 2016

Let’s place our first babysteps into reinforcement learning. I think, this one will be the very first part of a total of three parts series. In this series of posts, I would like to cover Q learning in reinforcement learning, starting from its pure implementation, through neural network based modification and finally adding computer vision facilities to our game playing system.

Oct 5, 2016

Hi, now I’m going to talk about my major project. You may have read my post regarding my mini project, which focused on the same goal of malayalam transliteration. In my previous attempt, I based it purely on basic probability theory. Which applied chain rule to transliterate to malayalam (literally). Now we’re going to adapt a state of the art technique which is very popular in this area, which is known as LSTM (Long Short Term Memory). Just before beginning my project, I’ve did a small research on it and found different papers. I’ve fallen in love with two simple papers, published one, by Google and the other by by Microsoft . No doubt, they have already proven the results. So, my project work is simply an implementation of these papers.

Jul 4, 2016

In natural language processing, when it comes to UNICODE text processing, it becomes slightly complex task to manage this unicode character, especially when it requires to randomly access characters from a sequence. Because of UNICODE’s very clever encoding, it is inherently sequential and a headache during random accesses. Possibilities are limited and we need error control and other corrective mechanisms to do so. Thus, most of the times, its a good idea to understand what our domain of unicode characters and map them into byte representations as ASCII characters do.

Jul 1, 2016

This quick reference that I’ve created for my own reference, is based on the book programming in lua - 3rd edition link.