Machine Learning (ML)

Wiener filter

Wiener filter performs two main functions - it inverts the blur of the image and removes extra noise. It is particularly helpful when processing images that have been through a degradation filter.

Summary: High Performance Convolutional Neural Networks for Document Processing

We have explored the paper "High Performance Convolutional Neural Networks for Document Processing" by Microsoft Research. It explores techniques to compute convolution layer in CNN faster.

Machine Learning (ML)

Understanding Deep Image Representations by Inverting Them

We have explore the problem that given an encoding of an image, to which extent is it possible to reconstruct the image itself? We have explored the paper "Understanding Deep Image Representations by Inverting Them".

Machine Learning (ML)

Overview of Semantic Segmentation

Semantic Segmentation is the process of labeling pixels present in an image into a specific class label. It is considered to be a classification process which classifies each pixel. The process of predicting each pixel in the class is known as dense prediction.

Machine Learning (ML)

Multinomial/ Multimodal Naive Bayes

Multimodal naive bayes is a specialized version of naive bayes designed to handle text documents using word counts as it's underlying method of calculating probability.

Machine Learning (ML)

Byte Pair Encoding for Natural Language Processing (NLP)

Byte Pair Encoding is originally a compression algorithm that was adapted for NLP usage. Byte Pair Encoding comes in handy for handling the vocabulary issue through a bottom-up process.

Machine Learning (ML)

Transformer Networks: How They Can Replace GANs

After an introduction to how transformers work, and a brief look at how they process text data, we see how they can also generate images and audio data just like GANs.

Machine Learning (ML)

“Show and Tell: A Neural Image Caption Generator” by Vinyals

This article explains the conference paper "Show and tell: A neural image caption generator" by Vinyals and others. This paper showcases how it approached state of art results using neural networks and provided a new path for the automatic captioning task.

Machine Learning (ML)

Solving Jigsaw Puzzles using Machine Learning

The paper "Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzles" describes a convolutional neural network (CNN) that aims to solve a pretext task, solving Jigsaw puzzles without manual labelling, and then to solve object classification and detection tasks.

Machine Learning (ML)

Graph Convolution Network (GCN)

Graph Convolution Network (GCN) are variants of Convolution Neural Network which brings in key ideas from Graph Theory. We have covered the key ideas of Graph Convolution Network (GCN) in depth.

Machine Learning (ML)

Automatic Image Annotation / Image Captioning

We have covered the general idea behind Automatic Image Annotation / Image Captioning and different techniques like retrieval based image captioning, template based and deep learning based image captioning.

Machine Learning (ML)

A Deep Learning Approach for Native Language Identification (NLI)

Native language identification (NLI) is the task of determining an author's native language based only on their writings or speeches in a second language. In this article, we will implement a model to identify native language of the author.

Machine Learning (ML)

Variants of Graph Neural Networks (GNN)

In this article, we have explore the different types of Graph Neural Networks (GNN) which is classified based on graph type, propagation step and training method.

Machine Learning (ML)

Gated Graph Sequence Neural Networks (GGSNN)

Gated Graph Sequence Neural Networks (GGSNN) is a modification to Gated Graph Neural Networks which three major changes involving backpropagation, unrolling recurrence and the propagation model. We have explored the idea in depth.

Machine Learning (ML)

Implementation of BERT

In this article I tried to implement and explain the BERT (Bidirectional Encoder Representations from Transformers) Model . It mainly consists of defining each component's architecture and implementing a python code for it.

Machine Learning (ML)

Advantages and Disadvantages of Linear Regression

Linear regression is a simple Supervised Learning algorithm that is used to predict the value of a dependent variable(y) for a given value of the independent variable(x). We have discussed the advantages and disadvantages of Linear Regression in depth.

Machine Learning (ML)

Capsule neural networks or CapsNet

Capsule neural networks or CapsNet is an artificial neural network which consists of capsules(bunch of neurons) which allow us to confirm if entities(components) are present in the image. We will cover Capsule Networks in depth.

Machine Learning (ML)

Overview of Generative Adversarial Networks (GANs) and their Applications

Today we cover Generative Adversarial Networks – or GANs for short. GANs are some of the most impressive things that we can do using deep learning (a sub-field of Machine Learning). We shall first see how they work, and then see some interesting and recent applications.

Machine Learning (ML)

Classification and Regression Trees: Advanced Methods (with C4.5 algorithm)

In this post, we show the popular C4.5 algorithm on the same classification problem and look into advanced techniques to improve our trees: such as random forests and pruning.

Machine Learning (ML)

Classification and Regression Trees (CART) Algorithm

Classification and Regression Trees (CART) is only a modern term for what are otherwise known as Decision Trees. Decision Trees have been around for a very long time and are important for predictive modelling in Machine Learning.

Machine Learning (ML)

Overview of Graph Neural Networks

Graph neural network is a special kind of network, which works with a graph as a data sample. The typical neural network works with arrays, while GNN works with graphs.

Machine Learning (ML)

John McCarthy, Man behind Garbage Collection

John McCarthy was one of the most influential Computer Scientists in 1950s and a Professor at Stanford University for nearly 38 years. He is best known for developing LISP Programming Language, inventing Garbage Collection, coining the word Artificial Intelligence and much more.

Machine Learning (ML)

Advantages and Disadvantages of Logistic Regression

In this article, we have explored the various advantages and disadvantages of using logistic regression algorithm in depth.

Machine Learning (ML)

Native Language Identification (NLI)

Native language identification (NLI) is the task of determining an author's native language based only on their writings or speeches in a second language. This is an application of Machine Learning.

Machine Learning (ML)

Disadvantages of CNN models

CNN (Convolutional Neural Network) is the fundamental model in Machine Learning and is used in some of the most applications today. There are some drawbacks of CNN models which we have covered and attempts to fix it.