Last Updated: 2/6/2023

Generative Computing

What is Generative Computing?

Generative computing is a field of computer science that focuses on creating systems that can generate new and unique outputs based on a set of inputs and rules. This can range from generating images and sounds to create entire systems or applications. Generative computing can be seen as artificial creativity, where the system creates new outputs based on rules and algorithms.

These AI-based generative models are trained on a large dataset of existing data and use that knowledge to generate new examples similar to the training data. Some examples of AI generative models include Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and Transformer-based models. These models can be used for many applications, such as image synthesis, text generation, music composition, etc. AI generative computing can generate new images, videos, and audio just as a human does, but it does not spot there. It can also create new drug molecules, new materials, and new solutions to scientific problems.

How Generative Computing Works

Generative computing systems work by taking a set of inputs and using algorithms and rules to create new outputs. These inputs can be anything from simple data to more complex information, such as music or images. The system's algorithms and rules determine how the inputs are transformed and combined to create the outputs.

One of the key characteristics of generative computing systems is that they can create new and unique outputs, even when given the same inputs. This means that the outputs produced by the system are not pre-determined but are generated based on the rules and algorithms used.

Classifications of Generative Computing

Generative computing is a subfield of Artificial Intelligence that focuses on creating new and original artifacts using computational processes, such as images, sounds, and text. There are several subfields within generative computing, including:

1. Generative Art: This subfield uses algorithms and computational processes to create new and original works of art. It encompasses various techniques, including generative graphics, generative music, and generative design.
2. Generative Design: This subfield uses algorithms to generate designs, such as architectural plans or industrial products. The goal of generative design is to automate the design process, making it faster and more efficient while also allowing the creation of new and innovative designs.
3. Generative Music: This subfield uses algorithms to generate music as complete pieces or as parts of compositions. Generative music can create new and original pieces or variations of existing works.
4. Generative Text: This subfield uses algorithms to generate text as complete pieces or as parts of larger works. Generative text can create new and original pieces, such as poetry or fiction, or to generate variations on existing works.
5. Generative Models: This subfield focuses on algorithms to model and generate data, such as images or sounds. Generative models can be used for tasks such as image synthesis, audio synthesis, and data augmentation.

The goal of generative computing is to automate the creation process, making it faster and more efficient while also allowing the creation of new and innovative artifacts.

Generative Computing in Real Life

• In the creative industry, generative art and music are becoming increasingly popular. Artists are using generative algorithms to create unique pieces of art and musicians are using them to compose new songs.
• In the fashion industry, generative design is being used to create unique and customizable clothing and accessories.
• In the gaming industry, generative algorithms are being used to create procedurally generated levels and game content.
• In the field of natural language processing, generative models are being used to generate human-like text, such as in chatbots, virtual assistants, and language translation systems.
• In the field of computer vision, generative models are being used to generate realistic images and videos, such as in facial recognition systems, and in virtual and augmented reality.
• In the field of drug discovery, generative models are being used to generate new drug candidates and to design new molecules with desired properties.
• In the field of machine learning, generative models are being used to generate synthetic data which can be used to improve model performance or to test models on unseen data.
• In the field of self-driving cars, generative models are being used to simulate different driving scenarios and test the capabilities of the car to react in a real-world situation.

Extra Resources

Books

Videos