Deep Learning
Deep Learning may be a subfield of machine learning involved with algorithms galvanized by the structure and performance of the brain known as artificial neural networks.
The leaders and specialists within the field have concepts of what deep learning is and these specific and nuanced views shed a great deal of sunshine on what deep learning is all regarding. during this post, you’ll discover precisely what deep learning is by hearing from a spread of consultants and leaders within the field.
Deep learning is a side of computer science (AI) that’s involved with emulating the training approach that individuals use to realize bound kinds of information. At its simplest, deep learning are often thought of as some way to automatise prophetical analytics.
Computer programs that use deep learning undergo abundant a similar method. every rule within the hierarchy applies a nonlinear transformation on its input and uses what it learns to make a applied math model as output. Iterations continue till the output has reached an appropriate level of accuracy. the amount of process layers through that information should pass is what galvanized the label deep.
Deep Learning:
Advanced Machine Learning:
Neural Networks Intro
- Artificial Neural Networks(ANN)
- Deep Neural Networks
- Convolutional Neural Networks(CNN)
- Recurrent Neural Networks(RNN)
- Stock Price Prediction using Neural Networks: Demo
- Neural Net Concepts:
- Neurons as Nodes: Perceptrons
- Dense & Sparse Neural Networks
Neuron Based approach: Benefits
- Perceptrons
- Learning Weights
- Gradient Descent & Back Propagation
- Activation Function & Feedforward Neural Networks
Installing Prerequisite Softwares:
- Tensorflow
- Theano
- Keras
3 Layer Neural Network for Customer Churn Modeling
Online Learning(Reinforcement Learning)
Generative Adversarial Networks (GANs)
PyTorch
Image Processing Introduction
- OpenCV for Image Processing in Python
- Edge Detection
- Eye & Nose Detection
- Face Detection using Haar cascades
- Optical Character Recognition using Neural Networks
- Text Detection: Sliding Window
- Character Segmentation
- Character Classification
Synthetic Character Generation: Shearing & Scaling, Rotation
Revisiting Perceptrons
- Coding a Text Classifier in Neural Networks
Advanced Neural Nets:
Long Short Term Memory(LSTM) in RNN
Time Series Data(ARMA, ARIMA)
Unsupervised Learning using Hidden Markov Model(Tensorflow and Theano)
Tensorflow Deep Dive
Speech Recognition
Advanced Text Mining
Building & Deploying a Intelligent Chatbot
- Data Preprocessing
- Seq2Seq
- Deploying the Chat Application
Computer Vision as AI
Image Recoginition and Classification
Deep Neural Networks Architecture revisited
Deep Convolutional Neural Network for Image Recognition:
- Convolutions
- Pooling, Flattening
- LeNet, Fully Connectected Feed Forward Network
- Face Recognition using Convolutional Neural Network
- Importing Pretrained Models
- Running Convolutional Neural Networks on GPU for Image
Unsupervised Learning in Deep Neural Networks Revisited
- Current Advancements:
- Self Organizing Maps(SOM)
- Auto Encoders
- Boltzman Machines
VGG, SDD, ResNet
Future Direction: Self Driving Cars, IIoT with AI, Drone based Parcel Delivery, etc
Conclusion
