How many times have you heard the words Artificial Intelligence (AI) today?
Did you realize that AI isn’t just one technique or method?
Did you realize you have already used and come in contact with multiple AI algorithms today?
Welcome to the first in a series of posts from the Veritas Automata Intelligent Data Team. Our Intelligent Data Practice helps you to understand your data and create solutions to leverage it, super powering your business.
Did you realize that AI isn’t just one technique or method?
Did you realize you have already used and come in contact with multiple AI algorithms today?
Welcome to the first in a series of posts from the Veritas Automata Intelligent Data Team. Our Intelligent Data Practice helps you to understand your data and create solutions to leverage it, super powering your business.
We are going to start by diving into some core definitions for Artificial Intelligence (AI) and Machine Learning (ML) in this introduction. Next we will expand on the core concepts, so you can learn how our team thinks and how we apply the right technology to the right problem in our Veritas Automata Intelligent Data Practice.
But it’s all just AI, right?
Well, yes and no. You could use that for general conversation purposes but if you are selecting a tool to solve a specific business challenge you will need to have a more fine-grained understanding of the space.
As Veritas Automata, we break AI down into two general categories:
01. Machine Learning (ML)
- We use Machine Learning to define algorithms that can be provable and deterministic (always return the same answer with the same data).
- Some example of techniques that fit in this space:
- Supervised Learning: Trains a model with labeled data to make predictions, like classifying medical images for diagnosis or assessing loan risk. Example: Image classification in healthcare or assessing risk for loans
- Unsupervised Learning: Finds patterns in unlabeled data; useful for spotting unusual behavior in fraud detection. Example: fraud detection
- Reinforcement Learning: A model learns by trial and error, such as a smart thermostat that adjusts to your preferred temperature and schedule to save energy. Example: Smart thermostat – as you adjust the temperature over time it learns your ideal temperature and when you are home/at work, it leverages this to optimize your home’s temperature and power usage.
- We use Generative AI to cover non-deterministic techniques, such as Generative Pre-trained Transformers (GPTs) and Large Language Models (LLM/GPTs), and Generative Adversarial Networks (GANs). These are both powerful techniques to create new content from existing data but the results are not 100% repeatable.
After we have covered the basics to set a baseline of what they are, we will do a deep dive of when you should choose which family of tools.
And lastly we will have deep dives into:
- The impact of copyright and ethics around GenAI
- The hybrid future of ML and GenAI
- Why you shouldn’t be afraid of AI and how it can help augment your career
Click below to continue with our Thought Leadership
Traditional Machine Learning –
Learning from Data
Learning from Data
