Neural Networks and Deep Learning
Recently, I have completed the course Neural Networks and Deep Learning as a part of deeplearning.ai specialization. By this article, I will try my best to articulate week 1 of the course.
Every new learner of deep learning always have a thought of
How neural network actually works?🤔
Is there any magic or what !!
May be bunch of if/else? 🤥
and yes, these type of questions must have to come in order to take deep dive in any learning. After completing this course, one have a gist of knowledge about working of neural networks, different structures, hidden layers, neurons, activation functions, learning rate and many more.
Week 1 : Introduction to Deep Learning
Starting with little introduction of specialization, I got to know about major trends which driving the rise of deep learning. Andrew Ng quoted,
AI is The New Electricity
He compares Electricity with Artificial Intelligence in the sense of growth and adaptability of AI in every industry.
After introduction the actual learning is started. Firstly, By taking example of house price prediction, I get to know about input layer, hidden layers and output layer. Input layer is not counted as layer in Neural Network. In above model there is total 2 layers. Input layer with 4 inputs, 1st layer is hidden layer with 3 neurons and 2nd layer is output layer with one neuron.
Then explored applications of supervised learning and learned which Neural Network works best for specific application.
Understand what are the major categories of models (such as CNNs and RNNs), and when they should be applied.
- CNN (Convolutional Neural Network) - Best to use when work with Image Data
- RNN (Recurrent Neural Network) - Gives best results when working with sequence data or temporal data
And finally, at the end of this curious first week, Andrew Ng has conducted a 40 minutes podcast type interview with Geoffrey Hinton, The Godfather of Deep learning.
So, This is the overview of first week of course Neural networks and deep learning. If some content were missed by me, then please let me know in the comments or you can direct message me. This is my first attempt to write such content so please give feedback. I would like to create more content about this type of overview of courses, projects and technical concepts. I will give my best try to express whole learning in simplest and shortest way.
Github Profile : 11fenil11
Thank You