Basics of machine learning and deep learning and its application in various fields.

1. 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.
2. This repository contains a list various tasks that can be achieved with Machine Learning/Deep Learning.
3. 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.

Theory:

Scikit Learn is a wonderful library for implementing basic machine learning tasks.

Naive Bayes:

• Basics of naive bayes with term frequency and inverse document frequecy intuition [site]

Deep Learning:

• 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]

Text Embedding:

• Word2vec [paper]
• Doc2vec [paper]
• SkipThought Vectors [paper] [implementation]

Applications:

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]