Foundations of Word Embeddings

Explain why representing words as dense vectors is advantageous over one-hot encoding.

Describe the skip-gram and CBOW architectures of Word2Vec.

Interpret geometric relationships in embedding spaces, such as word analogies via vector arithmetic.

Compare Word2Vec, GloVe, and fastText in terms of training approach and strengths.

Load and use pre-trained word embeddings to compute word similarities.