3
$\begingroup$

The paper Attention Is All You Need describes the transformer architecture that has an encoder and a decoder.

However, I wasn't clear on what the cost function to minimize is for such an architecture.

Consider a translation task, for example, where give an English sentence $x_{english} = [x_0, x_1, x_2, \dots, x_m]$, the transformer decodes the sentence into a French sentence $x_{french}' = [x_0', x_1', \dots, x_n']$. Let's say the true label is $y_{french} = [y_0, y_1, \dots, y_p]$.

What is the object function of the transformer? Is it the MSE between $x_{french}'$ and $y_{french}$? And does it have any weight regularization terms?

$\endgroup$
2
3
$\begingroup$

I took a look at the Tensor2Tensor's source code implementation as per @nbro's suggestion, and it seems like the loss function is the cross entropy between the predicted ($\|sentence length\|$ x $\|vocab\|$) probability matrix (right before taking the argmax to find the token to output), and the $\|sentence length\|$-length vector of token IDs as the true label.

$\endgroup$
2
  • $\begingroup$ Note that they reshape the out_logits variable there, in the line that you're linking us to, i.e. tf.reshape(out_logits, [-1, VOCAB_SIZE]), so the cross-entropy seems to be computed between two vectors and not a matrix and a vector (but I didn't really executed that code and tried to output the shape of those tensors/variables). It may be a good idea to do it, just to confirm the exact shape. $\endgroup$ – nbro Dec 10 '20 at 10:21
  • 1
    $\begingroup$ @nbro I just ran the code, and can confirm that is the case. out_logits originally has the shape (3, 7, 1, 1, 10), which is to (BATCH_SIZE, SENTENCE_LENGTH, ??, ??, VOCAB_SIZE). The ?? are bookkeeping dimensions because the tensor2tensor model expects a 5-D tensor, but are then immediately stripped out afterwards. Then, tf.reshape(out_logits, [-1, VOCAB_SIZE]) resizes out_logits to (21, 10), which is (BATCH_SIZE * SENTENCE_LENGTH, VOCAB_SIZE). Btw, for tf.reshape, seems like adding a -1 as a dimension makes it a flexible sized dimension, so it results in a 2D matrix rather than a 1D vector. $\endgroup$ – user3667125 Dec 13 '20 at 0:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.