## Grokking Deep Learning summary

Artificial intelligence is one of the most exciting technologies of the century, and deep learning is in many ways the “brain” behind some of the smartest artificial intelligence systems in the world. Freely based on the behavior of neurons inside the human brain, these systems are quickly catching up with the intelligence of their human creators, beating the world champion of Go, achieving superhuman performances in video games, driving cars, translating languages and sometimes even helping the police. fight against crime. Deep Learning is a revolution that is changing all industries around the world. Purchase of the print book includes a free eBook in PDF, Kindle and ePub format from Manning Publications.

### On technology

Deep learning, a branch of artificial intelligence, teaches computers to learn using neural networks, a technology inspired by the human brain. Online text translation, self-driving cars, personalized product recommendations, and virtual voice assistants are just some of the exciting modern advances made possible by deep learning.

### About the book

Grokking Deep Learning teaches you to build neural networks for deep learning from scratch! In his immersive style, deep learning expert Andrew Trask shows you the hidden science so you can uncover every detail of neural network training for yourself. Only with Python and its math-assisted NumPy library will you train your neural networks to see and understand images, translate texts into different languages, and even write like Shakespeare! Upon completion, you will be well prepared to master deep learning frameworks.

### What's inside?

- The science behind deep learning
- Development and training of own neural networks
- Data protection concepts, including federated learning
- Tips for continuing your deep learning research

### On the player

For readers with intermediate math and programming skills.

### About the author

Andrew Trask is a PhD student at the University of Oxford and a researcher at DeepMind. Andrew previously worked as an analytics researcher and product manager at Digital Reasoning, where he trained the world's largest artificial neural network and helped develop the analytics roadmap for Synthesys' cognitive computing platform.

## Andrew Trask Grokking Deep Learning

- Introduction to deep learning: why you should learn it
- Basic Concepts: How Do Machines Learn?
- Introduction to neural prediction: direct propagation
- Introduction to neural learning: gradual descent
- Learning multiple weights at the same time: generalizing the descent of the gradient
- Building your first deep neural network: introduction to backpropagation
- Here's how you imagine neural networks: in your head and on paper
- Learning the signal and ignoring the noise: introduction to regularization and dosage
- Probability and non-linearity modeling: activation functions
- Neural learning on edges and corners: introduction to convolutional neural networks
- Neural networks that understand language: King - Male + Female ==?
- Neural networks that write like Shakespeare: recurring layers for variable-length data
- Introduction to automatic optimization: let's create a deep learning framework
- Learn to write like Shakespeare: long short term memory
- Deep learning on invisible data: introduction of federated learning
- Where to go from here: a short guide