- True understanding is making all parts of a concept so clear and vivid as if they are objects in front of you that you can manipulate as you wish.
- This repository contains a list various tasks that can be achieved with Machine Learning/Deep Learning.
- It doesn't contain any benchmark paper because I love in doing new things rather than improving previous research.
Pull Requests are welcome to add any new blog, paper, site link.
Scikit Learn is a wonderful library for implementing basic machine learning tasks.
- Basics of naive bayes with term frequency and inverse document frequecy intuition [site]
- A beginners guide to understand neural nets. Easy explanation [book]
- Detailed explanation with math [book]
- Dropouts [paper]
- Practical approach and code in python. [Theano] [Tensorflow] [keras]
- RNNs and LSTMs Intuition [site]
- GANs [first-paper]
- Word2vec [paper]
- Doc2vec [paper]
- SkipThought Vectors [paper] [implementation]
Keyword Spotting system / Wake-word detection
- Small-footprint keyword spotting using deep neural networks [paper]
- Convolutional Neural Networks for Small-footprint Keyword Spotting [paper]
- A hidden Markov model based keyword recognition system [paper]
Voice activity detection
- A simple but effective real-time voice activity detection algorithm [paper]
- Recurrent neural networks for voice activity detection [paper]
Speech recognition and acoustic modeling
- Training LVCSR systems on thousands of hours of data [paper]
- Applying CNN on hybrid NN-HMM model for speech recognition [paper]
- Convolutional Neural Networks for Speech Recognition [paper]
- Long Short-Term Memory Recurrent Neural Network Architectures for Large Scale Acoustic Modeling [paper]
- Understanding how deep beleif networks perform acoustic modelling [paper]
Activity Detection
- Identifying Types of Physical Activity With a Single Accelerometer: Evaluating Laboratory-trained Algorithms in Daily Life [paper]
- A Study on Human Activity Recognition Using Accelerometer Data from Smartphones [paper]
- A Practical Approach to Recognizing Physical Activities [paper]
Face-recognition using FPGA
- A self-configurable systolic architecture for face recognition system based on principal component neural network [paper]
Text Classification
- Support Vector Machines and Word2vec for Text Classification with Semantic Features [paper]
Time Series Analysis
- An Introductory Study on Time Series Modeling and Forecasting [paper]
Retinal vessel segmentation
- Retinal Vessel Segmentation Using Deep Neural Networks [paper]
- Segmenting Retinal Blood Vessels With Deep Neural Networks [paper] [implementation]
- Retinal Vessel Segmentation Using the 2-D Gabor Wavelet and Supervised Classification[paper]
Text to Image Generation
- Text to image Synthesis using GANs [paper] [implementation]
- StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks [paper] [implementation]
- StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networks [paper] [implementation]
- AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networks [paper]
Object Classification and Recognition
- YOLO9000: Better, Faster, Stronger [paper] [implementation]
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks [paper] [implementation]
Neural Talk 2
- Image Captioning by Andrej Karpathy [implementation]
Neural VQA
- Exploring Models and Data for Image Question Answering [paper] [implementation]
Genre Classification
- To classify songs based on genre automatically [implementation]