Page 170 - AI Computer 10

The block representation of a Deep Learning model is shown below:

Input Output

Artificial Neural Network (ANN)

The key characteristics of deep learning include:
u Neural Networks: DL model uses multi-layered architectures known as neural networks to extract features
from data automatically, requiring little manual engineering.
u Large Datasets: DL models typically perform best when trained on vast amount of labeled data, leveraging
complex architectures to learn representations of the data.
u High Computational Power: Training DL models often requires significant computational resources, such as
GPUs, to handle the extensive matrix calculations involved.
Some examples of application of Deep Learning are:

u Object Identification and Object Localisation: Object Identification refers to a group of related computer
vision tasks that involve identifying objects in digital photographs. The objective of object classification is to
predict the class of one object in an image. Object Localisation refers to identifying the location of one or
more objects in an image and drawing a bounding box around their extent.

u Digital Recognition: Digital Recognition refers to the use of neural networks to recognise various forms of
data such as images, audio, and text. This includes tasks such as image recognition, speech recognition, and
handwriting recognition.

36
36