Machine Learning

Transfer Learning in NLP

Transfer learning is undoubtedly the new (well, relatively anyway) hot thing in deep learning right now. In vision…

Evaluation Methods for Dialog Systems

Spoken Dialog Systems (SDS) have become very popular recently, especially for goal completion tasks on mobile devices. Also, with the increasing use of IoT devices and their associated assistants like Alexa, Google Home, etc., systems that can converse with users…

An Introduction to Speech Recognition using WFSTs

Until now, all of my blog posts have been about deep learning methods or their application to NLP. Since the last couple of weeks, however, I have started learning about Automatic Speech Recognition (ASR). Therefore, I will also include…

How to obtain Sentence Vectors?

In several of my previous posts, I have discussed methods for obtaining word embeddings, such as SVD, word2vec, or GloVe. In this post, I will abstract a level higher and talk about 4 different methods that have been proposed to get embeddings for sentences.

Sparsity in Online Learning with Lasso Regularization

Sparse vectors have become popular recently for 2 reasons:

1. Sparse matrices require much less storage since they can be stored using various space-saving methods.
2. Sparse vectors are much…

A short note on Gradient Descent optimization algorithms

Irony detection in tweets

There was a SemEval 2018 Shared Task on “irony detection in tweets” that ended recently. As a fun personal project, I thought of giving it a shot, just to implement some new ideas. In this post, I will describe my approach for the problem along with some code.

Introduction to Learning Theory — Part 1

One of the most significant take-aways from NIPS 2017 was the “alchemy” debate spearheaded by Ali Rahimi. In the wake of the event, I have been trying to learn more about statistical learning theory, even though the concepts may not be readily applicable to…