Artificial neural networks (ann). Anns mimic the operations of the human brain to analyze and respond to received information. To do this, they use artificial neurons, which are activated according to a function that defines them, and are divided into several layers. These layers are divided into three groups: input, hidden, and output layers. There are very diverse anns with different characteristics. An interesting method used in some ann techniques is backpropagation . This method uses the error of the output obtained in the previous iteration as input in the next iteration, thus adjusting the output of the neurons. In this way, you can improve your answer by having information
about the mistake you have made in the past. Just like a person, they have the ability to learn from their mistakes . Deep learning or deep e commerce photo editing service learning (dl) represents an important development in the field of neural networks and has achieved great success in application to real-world problems. Dl is considered as the ability to exploit the representation of hierarchical features learned only from the available data. In short, they are very promising in extracting both features and patterns from complex data. Some examples of its use are: computer vision, automatic speech recognition and the recognition of audio and music signals. They have been shown to produce cutting-edge results
on various tasks. How machine learning and deep learning work computer vision or computer vision is a technology that aims to reproduce the effect of human vision by electronically perceiving and understanding an image. The process can be divided into three steps: detection, including segmentation and feature extraction; tracking routes ; and action comprehension including classifier and action label. How does a drone learn?