Updated: Jun 1, 2020
I have never given algorithms much thought until recently by stumbling across the Optometrist Algorithm. This algorithm facilitates fusion energy. If you haven't had the chance to look at the article, I suggest you do so, to have all the information needed for this sequencing blog post.
Now, let's talk shortly about what an algorithm is. An algorithm is a finite set of instructions to perform computations. Thinking in the scope of computer science, you can look at algorithms as a coded recipe that will execute once it runs into a trigger. There are more complex algorithms, however, the basic principle remains the same. In this article, we'll look at the Optometrist Algorithm. It signifies a new approach to solving highly unimaginably complex problems about plasma behavior. Plasma behavior is challenging to predict, generating vast amounts of data and kinds of variables.
Furthermore, engineers and scientists have to take the highly-nonlinear and temporally-varying interaction between the plasma, external controls, and its environment into account, additionally adding to the already complex experiment. All of these challenges and the complexity of the entire undertaking led scientists and engineers from TAE to reach out to Google to help them solve an issue. They needed some advanced computing network to help them with all the data and to make computations much quicker and to sort through a multitude of simulated settings that potentially keep the plasma under control. In essence, they needed a solution to optimize the performance of their plasma fusion experiments.
Optometrist Because It Needs the Human Eye
The solution to this complex problem came in the form of an algorithmic method called stochastic perturbation approximation. This method for optimizing systems with multiple unknown parameters was combined with human choice, hence the name "Optometrist." Correspondingly to getting an eyeglass prescription, the Optometrist Algorithm faces a human plasma physics expert with two alternative experimental settings and associated outcomes, the human, however, decides which one produces better results. There is an excellent video on how this works; let's check it out:
If you are interested in the paper written about this specific algorithm - here you go: Achievement of Sustained Net Plasma Heating in a Fusion Experiment with the Optometrist Algorithm. Furthermore, I find this article very informative.
If you are interested in reading more about algorithms, how they work, what types of algorithms there are - here is another great article.
Have a good one, and as always: stay curious!