Tricky question. In my experience is better to just look for math resources on classic upsampling method, since deep learning papers and books tend to give them for granted, or not something related to AI (they are after all analytic methods). Another reason is probably that the math is not that hard, and already the wikipedia pages offer a good description of the basic methods
(nearest-neighbours, bilinear interpolation, bicubic interpolation).
For some others interesting and more advance techniques I found useful this paper: Mathematical Techniques for Image Interpolation
even though it's not exhaustive since it doesn't mention well known alternatives to bilinear or bicubic upsampling like Lanczos, so for completeness I would read also A Study of Image Upsampling and Downsampling Filters
Regarding the deep learning part, it's more valuable to search for papers that studies not the upsampling methods themselves, but the artifact they introduce, especially related to tasks like super-resolution. As a starting point I would dig into this blogpost and its references.