Shivansh Mundra

Shivansh Mundra

Pre-Final Year Undergraduate Student

Indian Institute of Technology, Kharagpur

About Me

Hello! My name is Shivansh Mundra. I am a pre-final year undergraduate student at Indian Institute of Technology, Kharagpur.

Extrovert who is in love with Mathematics, Machine Learning and Philosophies ❤️.

I usually waste my time thinking ideas that could impact mass and try to build upon those. But often ending up watching Netflix because optimistic nihilism.

If you have anything to discuss about Artificial Intelligence, Books, Badminton, Computer Science, shoot a mail. You won’t regret!.

I am not who I think I am. I am not who you think I am. I am who I think you think I am.

Recent Posts

Intro to Imitation Learning: From Why to How!

Jude is an amateur football player but a wannabe great player. Jude is a die-hard fan of Christiano Ronaldo and watches every time he is playing anywhere. Jude wondered if he could exactly be like Ronaldo.
Apr 19, 2020 7 min read Technical, Artificial Intelligence

Multi Label Classification using Vowpal Wabbit library: From Why to How

What is Multi-Label classification? Think of a recent article/tweet you read and if you were asked to tag it! Probably you will come up with tags such as technology, politics, fashion, etc.
Apr 16, 2020 4 min read Technical

Paper Summaries

My notes of papers I read

Multi Modal Variational AutoEncoders

Why we have different representation for different domain, why can’t we have shared or single representation of all modalities like image, text and speech. This paper is a baby step in this direction. Variational Auto Encoders for Multi Modal Generative Network.
Aug 18, 2020 Artificial Intelligence, Technical

Understanding Deep Learning requires Re-Thinking Generalisation

Paper Link Generalisation in Machine Learning Models In Supervised Machine Learning, we train the model on training set and evaluate the model on a validation/testing set which is unseen during the training process.
Jun 22, 2020 Artificial Intelligence, Technical

Recurrent Multimodal Interaction for Referring Image Segmentation

This idea of treating this problem as sequential problem is innovative and having an mLSTM cell to encode both visual and linguistic features is also good. This gives model ability to forget all those pixel which defy correspondence initially. My guts are that this kind of model would work good even where there are small objects because at each time step there would be a reduction/change of probable pixels for segmentation.
Apr 29, 2020 Artificial Intelligence, Technical

My Journey to Natural Language Processing

Natural Language Processing I consider the reader to be familiar with normal machine learning terminology and methods. We here are going to deal with Supervised learning mostly. Since all machine learning models are mathematical function approximators, we can;t input a sentence to a mathematical model!
Apr 25, 2020 Technical, Artificial Intelligence

Xception: Deep Learning with Depthwise Separable Convolutions

We know each channel in Convolution block represents some high level feature say in human, each channel would map a body part(just for example!). Then this idea of depthwise conv proposes that in process of identifying one body part don’t process info from other body part info, it would disturb the signal in processing.
Apr 22, 2020 Technical, Artificial Intelligence

Popular Topics

AGI AutoEncoders Behaviour Cloning Classification CNN Computer Vision GAN ImageNet Imitation Learning LSTM Machine Learning Multimodal NLP Optimization PyTorch Reinforcement Learning Segmentation

Contact

A lot can happen over Mail!