/AI-papers

ML/DL papers which I found useful. Lots of them.

AI-papers

Base Repository cloned from https://github.com/rupak-118/AI-papers

This repository contains a list of papers as discussed on my blog series (23 Deep Learning Papers To Get You Started) on Medium.

It also contains additional research papers which I keep finding useful along my DL journey.

Part 1

  1. [TO READ] - A Few Useful Things to know about Machine Learning - Pedro Domingos : A very useful paper summarizing all the key learnings from an ML standpoint

  2. Introduction to CNNs : An article, pretty different from a conventional paper, but provides a comprehensive explanation through mathematical concepts and derivations of basic CNN elements

  3. Visualizing and Understanding CNNs - Zeiler and Fergus : Landmark paper in the history of deep learning. This opened up the doors to building more interpretable and easily explained deep learning models

Part 2

  1. Xavier initialization - Glorot, Bengio : Talks about the importance of appropriate weight initialization in neural networks

  2. Delving Deep into Rectifiers - Surpassing Human Level Performance on ImageNet Classification : Introduces PReLU activation function and He initialization (another weight initialization scheme)

  3. BatchNormalization : Talks about internal covariate shift and accelerating the training process of a deep neural network by reducing it

  4. Overview of Gradient Descent Optimization Algorithms : Explains the math and intuition behind different optimization algorithms coming under the Gradient Descent umbrella

  5. Dropout - Hinton et al. : Introduces the hugely popular regularization technique, Dropout. Dropout has been a key component in most neural network architectures since, and has also been highly successful in competitions.