Deep Learning Online Training

1

• DEEP LEARNING

2

• The following topics will be covered in our
• Deep Learning 
• Online Training

3
Deep Learning with TensorFlow

• Deep Learning is a subfield of machine learning
  concerned with algorithms inspired by the
  structure and function of the brain called
  artificial neural networks. It is intersection of
  statistics, artificial intelligence, and data to
  build accurate models. TensorFlow is one of the
  newest and most comprehensive libraries for
  implementing deep learning. With deep learning
  going mainstream, making sense of data and
  getting accurate results using deep networks is
  possible.

4
How it works

• A deep learning model is designed to continually
  analyze data with a logic structure like how a
  human would draw conclusions. To achieve this,
  deep learning uses a layered structure of
  algorithms called an artificial neural network
  (ANN).

5
What you will learn from this course

• This course will offer you an opportunity to
  explore various complex algorithms for deep
  learning. You will also learn how to train model
  to derive new features to make sense of deeper
  layers of data. Using TensorFlow, you will learn
  how to train model in supervise and unsupervised
  category.

6
Introduction to Deep learning

• AI and Deep learning
• Advantage of Deep learning
• Deep Learning Primitives
• Deep Learning Architecture
• The Neural viewpoint
• The Representation Viewpoint

7
TensorFlow Fundamentals
Introduction of Tensors Installation of Tensors Scalars, Vectors, and Matrices Matrix Mathematics Initializing Constant Tensors Basic Computation using TensorFlow Sampling Random Tensors TensorFlow Variable Tensor Addition and Scaling Matrix Operation Tensor Shape Manipulation Tensor Types TensorFlow Graphs TensorFlow Sessions Logistic Regression Model Building and Training
8
Introduction to Neural Network
Basic Neural Network The Neurons Single Hidden Layer Model Multiple Hidden Layer Model Input, Output, Hidden Layers Details of Activation Functions Sigmoid Function Hyperbolic Tangent Function, SoftMax Selection of Right Activation Functions Network learning technique Weight initialization Forward Propagation Backpropagation Optimization Algorithms Regularization
9
Linear and Logistic Regression with TensorFlow

• Overview of Linear and Logistic Regression
• Loss Functions
• Gradient Descent
• Automatic Differentiation Systems
• Learning with TensorFlow
• Training Linear and Logistic Regression model
• Evaluating Model Accuracy

10
Convolutional Neural Networks

• Visual Cortex Architecture
• Convolutional Layer
• Filters
• Stacking Multiple Feature Maps
• TensorFlow Implementation
• Pooling/Subsampling
• Fully Connected Layer
• MNIST digit classification example

11
Recurrent Neural Networks
Recurrent Neurons Memory cells Input and Output Sequences Basic RNNs in TensorFlow Static Unrolling through Time Dynamic Unrolling through Time Handling Variable Length Input/output Sequence Training RNNs Creative RNNs Deep RNNs Distributing a Deep RNN Across Multiple GPUs The Difficulty of Training over many Time Steps
12
Reinforcement Learning
Policy Search Introduction to OpenAI Gym Neural Network Policies The Credit Assignment Problem Policy Gradients Markov Decision Process Temporal Difference Learning and Q-Learning Approximate Q-Learning and Deep Q-Learning
13
Prerequisites

• Basic understanding of linear algebra , calculus 
  and probability  are must for really
  understanding deep learning . It is expected that
  one has some knowledge or experience in basic
  Python programming skills with the capability to
  work effectively with data structures .
  Understanding how to frame a machine learning
  problem, including how data is represented will
  be an added advantage.

14
Who can attend

• Anyone who has coding experience with an
  engineering background or relevant knowledge in
  mathematics and computer science can take this
  session to get understanding of Deep learning.

15
(No Transcript)