Page 337 - Ai_417_V3.0_C9_Flipbook
P. 337
AE VAE
Latent Space Deterministic, fixed-dimensional Probabilistic, continuous latent space
Representation encoding of input data. representation, allowing for sampling of
data points.
Reconstruction Loss Minimises the difference between Same as AE but also includes a
the input data and its reconstructed regulariser to enforce a Gaussian-like
output. distribution in the latent space.
Handling Overfitting Can suffer from overfitting due to the Less prone to overfitting due to the
fixed encoding structure. probabilistic nature of the latent
space, which allows for smoother
generalisation.
Applications Image compression, denoising, feature Data generation, Unsupervised
extraction. Learning, anomaly detection.
Training Complexity Relatively simpler training process. More complex training process due to
the inclusion of regularisation terms and
sampling from the latent space.
Examples of Generative AI
Generative AI has many applications, from art and music to language and natural language processing.
Let's study some examples of how Generative AI is being used in various fields.
Art
Generative AI can create new artworks by learning styles from famous painters and generating novel pieces in
similar styles. For example:
• AI artists like "AI Portraits" and "DeepArt" have gained popularity for their ability to create visually stunning images.
• The Next Rembrandt project used data analysis and 3D printing to create a new painting in the style of Rembrandt.
Gen AI can be used for style transferring and portrait creation.
Object Style
Gen AI created an image with a given style
Portrait Creation
Style Transferring
Introduction to Generative AI 335

