Attention Is All You Need

Abstract

The dominant sequence transduction models are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train.

Key Findings

• 1Introduced the Transformer architecture based entirely on self-attention mechanisms
• 2Eliminated the need for recurrence and convolutions in sequence models
• 3Achieved state-of-the-art results on English-to-German and English-to-French translation
• 4Demonstrated significantly faster training compared to recurrent architectures
• 5Introduced multi-head attention and positional encoding concepts

Impact & Significance