Comment by nico
2 years ago
Fascinating. Thank you for the explanation
I wonder if the same can be done for images
Been playing with images lately and outputting jpegs. It’s amazing how the exact same base pixels, can generate so many different images, but the more noisy/random the images get, the bigger their jpg file size is, and conversely, the more photo-like the image, smaller the jpg size
I doubt it will work as well for image classification, because the statistical regularities in pixel values of two images depicting the same object can differ significantly. For example, the statistical regularities of pixel values depicting a black cat at night are more similar to those depicting a black dog at night than those depicting a white cat in the middle of the day, making it difficult to use the compressed statistical regularities to classify cats vs. dogs correctly.
I’m sure not all algorithms work for everything, but if it can group images by some sort of meaning, then it can definitely do categorization
In the case of text documents you might have similar situations, and the algorithm still works, even across languages
I guess without trying, we’ll never know