🪄 TLDR (too long; didn’t read)
Three machine learning types are based on human assistant levels and the type of training used in the machine learning process.
- Supervised
- Unsupervised
- Reinforcement learning
🚀 Long story
In simple terms, machine learning is training the machine to predict future results based on past data.
For that, we need lots of data to train the machine to get better predictions.
Based on training data type we can divide machine learning into three major categories.
✅ Supervised Learning
Firstly, the machine learning modal needs to be trained using the questing and answer format.
X map to Y
X - Input (question)
Y - Output (answer)
After some fine-tuning of the machine learning algorithm, the modal predicts the answers for a similar type of question without the correct answer.
There are two major types of supervised machine learning are
-
Regression
Prediction is related to numbers.
-
Classification
The prediction is limited to a set of predefined categories.
📉 Unsupervised Learning
In unsupervised learning, without knowing the input and output data, the machine learning model tries to find hidden patterns in the data.
There are three major types of supervised machine learning are
-
Clustering
Grouping the data based on similar behaviors.
-
Dimensionality reduction
Simplifying the data without losing too much information.
-
Anomaly/Novelty detection
Detecting new or unusual patterns among previous data.
🤖 Reinforcement learning
The machine learning modal learns through award and penalty.
In the process of training if the experience gives an award the modal tag that path is correct if the experience gives a penalty the modal tag that path is incorrect.
Sometimes the collective knowledge is based on experience that machine learning modal run relatively.
📝 Summary
In this post, we went through the main types of machine learning based on human assistants and the type of training while training the machine learning modal.
Happy coding/learning! 🙂