Convolutional Neural Networks (CNNs)

(Image credit: © Kaggle)

Content

  1. Introduction
  2. Key Differences Between Machine Learning And Deep Learning
  3. Deep Learning model classification
  4. Neural Networks
  5. Types of Neural Network Diagrams
  6. Key Points
  7. Applications of Deep Learning
  8. Understanding the Neural Network Jargon
  9. Learning Terminology
  10. Other Important Terms
  11. Activation function
  12. How does the network learn?
  13. Reference

Introduction

Convolutional Neural Networks (CNNs) are a class of deep learning models primarily used for image and video processing (computer vision) tasks. They excel at learning spatial hierarchies and capturing local patterns. CNNs consist of convolutional layers that apply filters to input images, enabling feature extraction. These filters detect patterns such as edges, textures, and shapes. Following convolutional layers, pooling layers are often used to down sample the data, reducing dimensionality while preserving important features. Finally, fully connected layers are employed for classification tasks, where the extracted features are combined to make predictions. CNNs have achieved breakthrough performance in various computer vision tasks, including image classification, object detection, and image segmentation.
Schematic diagram of a basic convolutional neural network (CNN) architecture (Image Credit: © Van Hiep Phung and Eun Joo Rhee)

References

  1. Convolutional Neural Network Tutorial
  2. Datacamp tutorial.
  3. Analyticsvidhya tutorials.
  4. Getting started with Keras, Tensorflow and Deep Learning.
  5. Introduction to Deep Learning
  6. What is deep learning?
  7. Deep learning architectures
  8. Hands on Machine Learning with Scikit-Learn, Keras, & TensorFlow, Aurelien Geron

Some other interesting things to know:

Portfolio section Content