In this chapter, we include two case studies. The first is the use of attention-based architectures for the task of neural machine translation. The second is the application of memory-augmented networks to the task of question answering.
More information can be found at: Deep Learning for NLP and Speech Recognition by Springer