Machine Learning

1 readers

1 users here now

Community Rules:

Be nice. No offensive behavior, insults or attacks: we encourage a diverse community in which members feel safe and have a voice.
Make your post clear and comprehensive: posts that lack insight or effort will be removed. (ex: questions which are easily googled)
Beginner or career related questions go elsewhere. This community is focused in discussion of research and new projects that advance the state-of-the-art.
Limit self-promotion. Comments and posts should be first and foremost about topics of interest to ML observers and practitioners. Limited self-promotion is tolerated, but the sub is not here as merely a source for free advertisement. Such posts will be removed at the discretion of the mods.

founded 1 year ago

MODERATORS

communick@academy.garden

[D]In transformer models, why is there a query and key matrix instead of just the product? (alien.top)

submitted 11 months ago by lildaemon@alien.top to c/machinelearning@academy.garden

12 comments fedilink hide all child comments

The only time that the query and key matrices are used is to compute the attention scores. That is $v_i^T \cdot W_q^T W_k v_j$ But what is used is the matrix $W_q^T W_k$. Why not just replace $W_q^T W_k$ with a single matrix $W_{qv}$, and learn the matrix that is the product of W_q^T W_k instead of the matrices themselves? How does it help to have two matrices instead of one? And if it helps, why is that not done when applying matrices between neuron layers?

Chatgpt tells me that the reason is that it allows the model to learn a different representation for the query and key. But because they are just dotted together, it seems to me that you can just use the original embedding as the query with no loss of generality.

you are viewing a single comment's thread
view the rest of the comments

[–] Affectionate-Fish241@alien.top 1 points 11 months ago (1 children)

dammit all of the answers are fkin terrible. Looks like the ai bots took over or everyone in this subreddit has become braindead since the blackout.

You obviously don't do W_q @ W_k. That's totally stupid.

What transformers do is (x_i@W_q) @ (x_j@W_k) where x_i and x_j are two tokens in the sequence. This is an interaction operation. This can't be precomputed. What you see noted in the papers is Q = x_i @ W_q, and K = x_j @ W_k.

(Transposes omitted for notational clarity, work that out yourself)

[–] mrfox321@alien.top 1 points 11 months ago

Your answer is also terrible. It does not answer his question.

Look at the top 2 replies to see correct interpretations of the question.