Skip to main content

Must Read Papers for Data Science, ML, and DL

Curated collection of Data Science, Machine Learning and Deep Learning papers, reviews and articles that are on must read list.โ€‹


๐Ÿ‘‡ READ THIS ๐Ÿ‘‡โ€‹

๐Ÿ‘‰ Reading paper with heavy math is hard, it takes time and effort to understand, most of it is dedication and motivation to not quit, don't be discouraged, read once, read twice, read thrice,... until it clicks and blows you away.

๐Ÿฅ‡ - Read it first

๐Ÿฅˆ - Read it second

๐Ÿฅ‰ - Read it third


Data Scienceโ€‹

๐Ÿ“Š Pre-processing & EDAโ€‹

๐Ÿฅ‡ ๐Ÿ“„Data preprocessing - Tidy data - by Hadley Wickham

๐Ÿ““ General DSโ€‹

๐Ÿฅ‡ ๐Ÿ“„ Statistical Modeling: The Two Cultures - by Leo Breiman

๐Ÿฅˆ ๐Ÿ“„ A study in Rashomon curves and volumes: A new perspective on generalization and model simplicity in machine learning

๐Ÿฅ‡ ๐Ÿ“„ Frequentism and Bayesianism: A Python-driven Primer by Jake VanderPlas


Machine Learningโ€‹

๐ŸŽฏ General MLโ€‹

๐Ÿฅ‡ ๐Ÿ“„ Model Evaluation, Model Selection, and Algorithm Selection in Machine Learning - by Sebastian Raschka

๐Ÿฅ‡ ๐Ÿ“„ A Brief Introduction into Machine Learning - by Gunnar Ratsch

๐Ÿฅ‰ ๐Ÿ“„ An Introduction to the Conjugate Gradient Method Without the Agonizing Pain - by Jonathan Richard Shewchuk

๐Ÿฅ‰ ๐Ÿ“„ On Model Stability as a Function of Random Seed

๐Ÿ” Outlier/Anomaly detectionโ€‹

๐Ÿฅ‡ ๐Ÿ“ฐ Outlier Detection : A Survey

๐Ÿš€ Boostingโ€‹

๐Ÿฅˆ ๐Ÿ“„ XGBoost: A Scalable Tree Boosting System

๐Ÿฅˆ ๐Ÿ“„ LightGBM: A Highly Efficient Gradient BoostingDecision Tree

๐Ÿฅˆ ๐Ÿ“„ AdaBoost and the Super Bowl of Classifiers - A Tutorial Introduction to Adaptive Boosting

๐Ÿฅ‰ ๐Ÿ“„ Greedy Function Approximation: A Gradient Boosting Machine

:book: Unraveling Blackbox MLโ€‹

๐Ÿฅ‰ ๐Ÿ“„ Peeking Inside the Black Box: Visualizing Statistical Learning with Plots of Individual Conditional Expectation

๐Ÿฅ‰ ๐Ÿ“„ Data Shapley: Equitable Valuation of Data for Machine Learning

โœ‚๏ธ Dimensionality Reductionโ€‹

๐Ÿฅ‡ ๐Ÿ“„ A Tutorial on Principal Component Analysis

๐Ÿฅˆ ๐Ÿ“„ How to Use t-SNE Effectively

๐Ÿฅ‰ ๐Ÿ“„ Visualizing Data using t-SNE

๐Ÿ“ˆ Optimizationโ€‹

๐Ÿฅ‡ ๐Ÿ“„ A Tutorial on Bayesian Optimization

๐Ÿฅˆ ๐Ÿ“„ Taking the Human Out of the Loop: A review of Bayesian Optimization


Famous Blogsโ€‹

Sebastian Raschka Chip Huyen


๐ŸŽฑ ๐Ÿ”ฎ Recommendersโ€‹

Surveysโ€‹

๐Ÿฅ‡ ๐Ÿ“„ A Survey of Collaborative Filtering Techniques

๐Ÿฅ‡ ๐Ÿ“„ Collaborative Filtering Recommender Systems

๐Ÿฅ‡ ๐Ÿ“„ Deep Learning Based Recommender System: A Survey and New Perspectives

๐Ÿฅ‡ ๐Ÿ“„ ๐Ÿค” โญ Explainable Recommendation: A Survey and New Perspectives โญ

Case Studiesโ€‹

๐Ÿฅˆ ๐Ÿ“„ The Netflix Recommender System: Algorithms, Business Value,and Innovation

๐Ÿฅˆ ๐Ÿ“„ Two Decades of Recommender Systems at Amazon.com

๐Ÿฅˆ ๐ŸŒ How Does Spotify Know You So Well?

๐Ÿ‘‰ More In-Depth study, ๐Ÿ“• Recommender Systems Handbook


Famous Deep Learning Blogs ๐Ÿค โ€‹

๐ŸŒ Stanford UFLDL Deep Learning Tutorial

๐ŸŒ Distill.pub

๐ŸŒ Colah's Blog

๐ŸŒ Andrej Karpathy

๐ŸŒ Zack Lipton

๐ŸŒ Sebastian Ruder

๐ŸŒ Jay Alammar


๐Ÿ“š Neural Networks and Deep Learning Neural Networksโ€‹

โญ ๐Ÿฅ‡ ๐Ÿ“ฐ The Matrix Calculus You Need For Deep Learning - Terence Parr and Jeremy Howard โญ

๐Ÿฅ‡ ๐Ÿ“ฐ Deep learning -Yann LeCun, Yoshua Bengio & Geoffrey Hinton

๐Ÿฅ‡ ๐Ÿ“„ Generalization in Deep Learning

๐Ÿฅ‡ ๐Ÿ“„ Topology of Learning in Artificial Neural Networks

๐Ÿฅ‡ ๐Ÿ“„ Dropout: A Simple Way to Prevent Neural Networks from Overfitting

๐Ÿฅˆ ๐Ÿ“„ Polynomial Regression As an Alternative to Neural Nets

๐Ÿฅˆ ๐ŸŒ The Neural Network Zoo

๐Ÿฅˆ ๐ŸŒ Image Completion with Deep Learning in TensorFlow

๐Ÿฅˆ ๐Ÿ“„ Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

๐Ÿฅ‰ ๐Ÿ“„ A systematic study of the class imbalance problem in convolutional neural networks

๐Ÿฅ‰ ๐Ÿ“„ All Neural Networks are Created Equal

๐Ÿฅ‰ ๐Ÿ“„ Adam: A Method for Stochastic Optimization

๐Ÿฅ‰ ๐Ÿ“„ AutoML: A Survey of the State-of-the-Art

๐Ÿ–ผ CNNsโ€‹

๐Ÿฅ‡ ๐Ÿ“„ Visualizing and Understanding Convolutional Networks -by Andrej Karpathy Justin Johnson Li Fei-Fei

๐Ÿฅˆ ๐Ÿ“„ Deep Residual Learning for Image Recognition

๐Ÿฅˆ ๐Ÿ“„AlexNet-ImageNet Classification with Deep Convolutional Neural Networks

๐Ÿฅˆ ๐Ÿ“„VGG Net-VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION

๐Ÿฅ‰ ๐Ÿ“„ A Mathematical Theory of Deep Convolutional Neural Networks for Feature Extraction

๐Ÿฅ‰ ๐Ÿ“„ Large-scale Video Classification with Convolutional Neural Networks

๐Ÿฅ‰ ๐Ÿ“„ Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering

โšซ CapsNet ๐Ÿ”ฑโ€‹

๐Ÿฅ‡ ๐Ÿ“„ Dynamic Routing Between Capsules

๐Ÿž ๐Ÿ’ฌ Image Captioningโ€‹

๐Ÿฅ‡ ๐Ÿ“„ Show and Tell: A Neural Image Caption Generator

๐Ÿฅˆ ๐Ÿ“„ Neural Machine Translation by Jointly Learning to Align and Translate

๐Ÿฅˆ ๐Ÿ“„ StyleNet: Generating Attractive Visual Captions with Styles

๐Ÿฅˆ ๐Ÿ“„ Show, Attend and Tell: Neural Image Caption Generation with Visual Attention

๐Ÿฅˆ ๐Ÿ“„ Where to put the Image in an Image Caption Generator

๐Ÿฅˆ ๐Ÿ“„ Dank Learning: Generating Memes Using Deep Neural Networks

:car: ๐Ÿšถ Object Detection ๐Ÿฆ… ๐Ÿˆโ€‹

๐Ÿฅˆ ๐Ÿ“„ResNet-Deep Residual Learning for Image Recognition

๐Ÿฅˆ ๐Ÿ“„ YOLO-You Only Look Once: Unified, Real-Time Object Detection

๐Ÿฅˆ ๐Ÿ“„ Microsoft COCO: Common Objects in Context

๐Ÿฅˆ ๐Ÿ“„ (R-CNN) Rich feature hierarchies for accurate object detection and semantic segmentation

๐Ÿฅˆ ๐Ÿ“„ Fast R-CNN

๐Ÿฅˆ ๐Ÿ“„ Faster R-CNN

๐Ÿฅˆ ๐Ÿ“„ Mask R-CNN

:car: ๐Ÿšถ ๐Ÿ‘ซ Pose Detection :runner: ๐Ÿ’ƒโ€‹

๐Ÿฅˆ ๐Ÿ“„ DensePose: Dense Human Pose Estimation In The Wild

๐Ÿฅˆ ๐Ÿ“„ Parsing R-CNN for Instance-Level Human Analysis

๐Ÿ”ก ๐Ÿ”ฃ Deep NLP ๐Ÿ’ฑ ๐Ÿ”ขโ€‹

๐Ÿฅ‡ ๐Ÿ“„ A Primer on Neural Network Models for Natural Language Processing

๐Ÿฅ‡ ๐Ÿ“„ Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling

๐Ÿฅ‡ ๐Ÿ“„ On the Properties of Neural Machine Translation: Encoderโ€“Decoder Approaches

๐Ÿฅ‡ ๐Ÿ“„ LSTM: A Search Space Odyssey - by Klaus Greff et al.

๐Ÿฅ‡ ๐Ÿ“„ A Critical Review of Recurrent Neural Networksfor Sequence Learning

๐Ÿฅ‡ ๐Ÿ“„ Visualizing and Understanding Recurrent Networks

โญ ๐Ÿฅ‡ ๐Ÿ“„ Attention Is All You Need โญ

๐Ÿฅ‡ ๐Ÿ“„ An Empirical Exploration of Recurrent Network Architectures

๐Ÿฅ‡ ๐Ÿ“„ Open AI (GPT-2) Language Models are Unsupervised Multitask Learners

๐Ÿฅ‡ ๐Ÿ“„ BERT: Pre-training of Deep Bidirectional Transformers forLanguage Understanding

๐Ÿฅ‰ ๐Ÿ“„ Parameter-Efficient Transfer Learning for NLP

๐Ÿฅ‰ ๐Ÿ“„ A Sensitivity Analysis of (and Practitionersโ€™ Guide to) ConvolutionalNeural Networks for Sentence Classification

๐Ÿฅ‰ ๐Ÿ“„ A Survey on Recent Advances in Named Entity Recognition from Deep Learning models

๐Ÿฅ‰ ๐Ÿ“„ Convolutional Neural Networks for Sentence Classification

๐Ÿฅ‰ ๐Ÿ“„ Pervasive Attention: 2D Convolutional Neural Networks for Sequence-to-Sequence Prediction

๐Ÿฅ‰ ๐Ÿ“„ Single Headed Attention RNN: Stop Thinking With Your Head

๐Ÿ‘ฝ GANsโ€‹

๐Ÿฅ‡ ๐Ÿ“„ Generative Adversarial Nets - Goodfellow et al.

๐Ÿ“š GAN Rabbit Hole -> GAN Papers

โญ•โž–โญ• GNNs (Graph Neural Networks)โ€‹

๐Ÿฅ‰ ๐Ÿ“„ A Comprehensive Survey on Graph Neural Networks


๐Ÿ‘จโ€โš•๏ธ ๐Ÿ’‰ Medical AI ๐Ÿ’Š ๐Ÿ”ฌโ€‹

Machine learning classifiers and fMRI: a tutorial overview - by Francisco et al.


๐Ÿ‘‡ Cool Stuff ๐Ÿ‘‡โ€‹

๐Ÿ”Š ๐Ÿ“„ SoundNet: Learning Sound Representations from Unlabeled Video

๐ŸŽจ ๐Ÿ“„ CAN: Creative Adversarial NetworksGenerating โ€œArtโ€ by Learning About Styles andDeviating from Style Norms

๐ŸŽจ ๐Ÿ“„ Deep Painterly Harmonization

๐Ÿ•บ ๐Ÿ’ƒ ๐Ÿ“„ Everybody Dance Now

โšฝ Soccer on Your Tabletop

๐Ÿ‘ฑโ€โ™€๏ธ ๐Ÿ’‡ ๐Ÿ“„ SC-FEGAN: Face Editing Generative Adversarial Network with User's Sketch and Color

๐Ÿ“ธ ๐Ÿ“„ Handheld Mobile Photography in Very Low Light

๐Ÿฏ ๐Ÿ•Œ ๐Ÿ“„ Learning Deep Features for Scene Recognitionusing Places Database

๐Ÿš… ๐Ÿš„ ๐Ÿ“„ High-Speed Tracking withKernelized Correlation Filters

๐ŸŽฌ ๐Ÿ“„ Recent progress in semantic image segmentation

Rabbit hole -> ๐Ÿ”Š ๐ŸŒ Analytics Vidhya Top 10 Audio Processing Tasks and their papers

๐Ÿ‘ฑ -> ๐Ÿ‘ด ๐Ÿ“„ ๐Ÿ“„ Face Aging With Condintional GANS

๐Ÿ‘ฑ -> ๐Ÿ‘ด ๐Ÿ“„ ๐Ÿ“„ Dual Conditional GANs for Face Aging and Rejuvenation

โš– ๐Ÿ“„ BAGAN: Data Augmentation with Balancing GAN

labml.ai Annotated PyTorch Paper Implementations


๐Ÿ“ฐ Cap Stone Projects ๐Ÿ“ฐโ€‹

8 Awesome Data Science Capstone Projects

10 Powerful Applications of Linear Algebra in Data Science

Top 5 Interesting Applications of GANs

Deep Learning Applications a beginner can build in minutes