'How does word2vec learn word relations?

Which part of the algorithm specifically makes the embeddings to have the king - boy + girl = queen ability? Did they just did this by accident?

Edit :

Take the CBOW as an example. I know about they use embeddings instead of one-hot vectors to encode the words and made the embeddings trainable instead of how we do when using one hot vectors that the data itself is not trainable. Then the output is a one-hot vector for target word. They just average all the surrounding word embeddings at some point then put some lego layers afterwards. So at the end they find the mentioned property by surprise, or is there a training procedure or network structure that gave the embeddings that property?

word2vecembeddingword-embedding


Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source

Related Questions

Ensure the gensim generate the same Word2Vec model for different runs on the same data

How to calculate similarity for pre-trained word embeddings

Read GloVe pre-trained embeddings into R, as a matrix

tensorflow2.x keras Embedding layer process tf.dataset error

Bert embedding layer raises 'ValueError: A target array with shape ' with BiLSTM in keras tensorflow

ReadError: file could not be opened successfully. But I am not sure where the tar file is stored to resolve this

How to store Bag of Words or Embeddings in a Database

How are the TokenEmbeddings in BERT created?

Calculating Similarity Between Pairs of Documents in R [closed]

Is there any faster way to get word embeddings given sub-word embeddings in BERT

Update an element in faiss index

gensim error : no module named gensim

Ensure the gensim generate the same Word2Vec model for different runs on the same data

How to load a pre-trained Word2vec MODEL File and reuse it?

How to calculate similarity for pre-trained word embeddings