Your distinction between TFIDF as simply computed vs embeddings as learned is odd and artificial. Both are computations from data, but TFIDF has an understandab

Your distinction between TFIDF as simply computed vs embeddings as learned is odd and artificial. Both are computations from data, but TFIDF has an understandable closed form and word embeddings do not. As for machine learning, it has to do with improvement and doesn't even necessarily need data.