🤖 Herbie’s Thoughts on Robotics (and Other Stuff)

The Ramblings of a (soon-to-be) Robotics PhD Student


Overfitting is Fine, Actually?

2025 May 12

Apparently, people in the machine learning community used to believe in something called the bias-variance tradeoff. The story went like this: if you use an ML algorithm that is too simplistic for your data, the algorithm will be too biased and perform poorly on both the train and test set. Fair enough. On the other hand, the story continues, if your algorithm is too expressive and complex… Read More →


Recent Posts


You can find my personal website here