AI + LLM courses specifically designed for entrepreneurs and senior professionals!!
In the first part of this article, we learned AI can’t do everything and rely mostly on previous data to learn.
Though it looks like AI can do everything yet there is a handful of activities only that AI models are capable of doing currently.
At the core, there are 3 main branches of ML
- Data science – Learns from tabular data. Data can be numeric or boolean or have multiple options etc.
- Computer Vision – Anything related to Image and Video
- Natural language processing (NLP) – related to text and language
Though these 3 branches look completely different, there is one thing which is common. The model needs lots of data to find a pattern and then applies it to the new unseen data.
For ex – in computer vision, if you want to predict a model that differentiates between a dog and a horse, it needs thousands of images of a dog and a horse.
NLP deals with languages and finds patterns in languages and for that also it needs previous data.
What are the current possibilities for AI?
I have just made the skeleton of this article and listed all the activities. Will describe all in detail in coming articles. Stay tuned for that.
Computer Vision –
- Image Classification
- Object detection in an image
- Multi-label classification in an image
- Image enhancement
Natural language processing
- Sentiment Analysis
- Question Answering
Multi-modal (deals with multiple modes, like Images and Text)
- Text to speech
- Speech to text
- Text to image
- Image to text
- Text to Video
- Image generation
- Text Generation
- Image to Video
- Video to Video
- Speech generation
- Code Generation
- Music Generation
- Drug discovery
- Speech Recognition
- Speech Enhancement
- Voice conversion
- Voice recognition
- Code Generation
- Text to SQL
- Motion Planning
- Robot Navigation
- Visual Navigation
- Music Information Retrieval
- Music Transcription
- Music modelling
This list might look long but it’s less than 1% of the things that humans can do or happens in real life.
There could across some more use cases but they are more or less extensions of the above models only.
I hope when next time, someone tells you about an AI solution, you can easily identify if it’s practically possible or not.
For a detailed list of State of the art AI models, check this
Thanks for choosing to read it completely. I can try to simplify any AI concept for you.
In case you want me to simplify any AI topic, please let me know on LinkedIn – https://www.linkedin.com/in/nikhileshtayal/