| tags | ||
|---|---|---|
|
Transformer is a revolutionary architecture for sequence-to-sequence or just sequence tasks, which was introduced by Vaswani et al., (2017).
The model is composed of two sub-models: encoder and decoder. Encoder recieves the input sequence, and makes the processed information available for the decoder. Decoder predicts the target sequence using past true target tokens (teacher forcing) and the input information provided by the encoder.
Encoder digests the input sequence via positional and semantic embedding matricies. For text this means that each sub-word get positional and semantic embedding. Though both embeddings can be trained, in the original paper only the semantic embedding matrix was trained, while the positional embeddings were computed. Then the input goes through several identical blocks. Each block consists of multi-head self-attention, and feed-forward layers.
Decoder is similar to encoder, yet not entirely the same. Decoder too ingests input through embeded tokens, which also pass through several identical layers. The layers are also similar to encoder's layers:
- self-attention
- encoder-decoder attention
- feed-forward layers.
There is a whole note about self-attention, so I mention only the specifics to its use. There are 3 usages of the attention mechanism:
- self-attention in encoder
- self-attention in decoder
- encoder-decoder attention
All types follow the basic computation but differ in some details. 1. is vanilla self-attention as it is described in the above mentioned note. 2. is the same, except its attention is masked out for upper diagonal to prevent the decoder accessing future tokens. The diagonal is not masked since the decoder's input is shifted by one. 3. is same as 1. except the values and keys are supplied by the processed sequence at the end of the encoder. This allows the decoder to ask for information from the input to generate the output.
Feed-forward layers include two feed-forward layers w/ ReLU between them which are applied to each token separatelly. The first layer scales the input dimension 4x, while the other scales it back down to the original size.
In the original article there were two residual connections that went:
- before the self-attention to after it
- before the feed-forward layers to after them Right after summing with the output of the avoided layer(s), there were layer-normalizations.
However, later it was discovered (as it was in ResNet in 2016) that putting the layer normalization inside the computation branch before any other computation is beneficial as the residual connection only contains sums and therefore doesn't disturb the gradient. The first usage of these pre-activation residual blocks in Transformers appears (AFAIK) in Sparse transformers paper from 2019.
During training the encoder is fed with inputs, decoder with shifted outputs and we use Cross-entropy loss on the decoder outputs.
During inference, decoding is done auto-regressively. Decoder predicts one token at a time, basing his next token prediction on the previous one.