The Ghibli Effect , Discovering Generative AI

As the Ghibli Art trends through Generative AI, let’s understand what is Generative AI, a subset of deep learning.

What is Generative AI ?

Generative AI = Generative + AI 
[Generating contents using artificial intelligence]

Generative Artificial Intelligence is a type of AI that can create new content, such as audio, images, text, code, videos etc.

Text Generation : AI models like ChatGPT, Bard, Claude generate coherent and contextually appropriate text based on the input they receive.

Image Generation : AI models like Midjourney, DALL.E can create images from text prompts, blending different concepts creatively.

Code Generation : Models like GitHub Copilot powered by OpenAI’s Codex, Copilot suggests code as you type, and can even write entire functions or classes based on comments or function names.

Understanding AI

Humans are the most intelligent form of species that has ever existed on earth. We excel in everything that we do. AI is a broad field of computer science that focuses on creating intelligent systems capable of performing tasks that typically require human intelligence.

For example, you can create AI systems that can diagnose diseases from x-ray, or they can predict property prices, or they can do editing for your videos.

Machine learning

Machine learning is a subset of AI that focuses on the development of algorithms and models that enable computers to learn and make predictions or decisions by learning on their own without explicit programming.

How machines learn? Same as human, machines also see lot of examples of datasets, memorise what it is, what are the unique attributes of that object, so that next time it will identify.

Deep Learning

Deep Learning is a subset of machine learning that focuses on teaching computers to learn and make decisions by processing data through neural networks which is inspired by the human brain.

Where does Generative AI fits ?

Generative AI is a subset of Deep learning. it uses deep learning neural networks to train models and generate new content.

Do u know how existing AI systems works ? Existing AI systems we call it as Conventional AI systems operates in a way that the model learn from the training data and then it either make predictions or classified the data into categories or did language processing like translations or computer visions like image identification.

Input data --> AI --> Predictions
                      Classifications
                      Computer Vision

Now will see how new AI that is Generative AI works ! Generative AI is entirely different from Conventional AI. Twice or trice the amount of data is been feed to the Generative AI, based on the training data, based on the neural network it is able to generate new contents, whether it is text or image or video.

2x(Input data) --> AI --> Generates new
                          Text
                          Image
                          Video

Here, it is generating a new image or video or anything using input data’s.

Conclusion

In order to train the machine to be more intelligent, more accurate it needs to be feed with large volume of data. Since we are feeding large volume of data, using neural networks, expecting quick answers when we type something into AI models, so the computational power has to be massive in order to process the data in very short amount of time. And a major shift to conversational and contextual interaction with AI Models are very popular these days. This helps the users to generate what they want, how they want quickly and easily.

---

Good Day ッ

Thanks for sticking with me!