= Machine learning as a form of data compression
{tag=Overfitting}
Philosophically, machine learning can be seen as a form of <lossy compression>.
And if we make it too <lossless>, then we are basically <overfitting>.
Bibliography:
* https://bair.berkeley.edu/blog/2019/09/19/bit-swap/
* https://www.eecs.tufts.edu/~dsculley/papers/compressionAndVectors.pdf
* https://arxiv.org/abs/2202.06533
* https://towardsdatascience.com/ai-based-image-compression-the-state-of-the-art-fb5aa6042bfa
Back to article page