I always promised myself I’d never be one of those people that starts a blog post with “Here’s why I haven’t posted in a while” but in this case, it matters.
I signed up the the Stanford Machine Learning course on Coursera, and it’s the most challenging course I’ve ever taken.
In most ways, that’s a good thing. The course forces you to learn the low-level under-the-hood mechanisms that power ML tool kits. I’ve built my own gradient descent algorithms, calculated loss, and manually plotted complex data. I’ve learned about contour graphs, minimization functions and matrix multiplication. …
The most interesting subset of machine learning to me has always been what I’ve learned to call Reinforcement Learning. The process of letting the software explore an environment and adapt to achieve the highest level of success.
For this first attempt, I began with this tutorial.
The goal is to teach the cart to reach the flag as efficiently as possible. A push in any single direction won’t do the job, so we must build a Q table that adapts to the reward from each attempt at swinging back and forth.
The environment provides three possible actions: Push Left, Inaction…
Wannabe machine-learning specialist