A team of scientists hailing from the Sandia National Laboratories and Boston University developed an experimental algorithm that could automatically diagnose problems in supercomputers. There is an array of internal and external issues that could arise with these powerful machines. For instance, factors like physical parts breaking can occur or previous programs performing "zombie processes" that prevent the computer from functioning properly. Furthermore, the repair process for these devices can take an extended period of time, which raises another issue since these computers perform critical tasks like forecasting the weather and ensuring the U.S. nuclear arsenal is safe and reliable without needing to do underground testing. To develop the algorithm, the team took a multi-step approach. First, the engineers created a suite of issues they became familiar with over the time they spent working on various supercomputers, which was then followed by them writing specific codes to re-create these anomalies.

When we use the 1st, 2nd, and nth (with a large n) order polynomial regression, we may get the following results. As we can see, the 1st order model is too simple to fit the data, the 2nd order model has reasonable fitting, and the nth order model is very complicated and can fit every single sample very well. But the problem with the 3rd case is that machine learning is not about fitting the training data, but about generalizing beyond the training data. While it can achieve small training error, it will have large test error.