Note
Ok, I’m obviously new to machine learning – and this post is mostly just for my reference – as a starting point for further learning and research.
Supervised and Unsupervised Machine Learning
Supervised to me just means you know what you are looking for. Unsupervised conversely means find trends in the data based on the data – not based on pre-conceived notions.
Supervised Learning
Within Supervised – Gradient Descent (a form of least squares curve fit) is widely used – I BELIEVE – research to confirm. Just derive a function f(x) = x where the function fits the data points as closely as possible.
More generally supervised machine learning has three major components:
- Hypothesis Function
- Loss Function
- Optimization Procedure
Statistics As It Relates To Machine Learning
Statistics = Probability + Data
Random Variables
Categorized Values or Variables – Finite range of values (e.g. small, medium, large )
Quantitavie Values or Variables – Continuous Values X in the domain of Real Numbers
Key terms in Machine Learning
Gradient Descent
First understanding the Gradient. Gradient is another word for derivative, or the rate of change of a function. It’s a vector (a direction to move) that. Points in the direction of greatest increase of a function.
TensorFlow
Reference:
https://web.stanford.edu/class/cs224n/