BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Modeling Emotional Dialogs with Sequence to Sequence Networks
DTSTART:20180827T090000
DTEND:20180827T110000
DTSTAMP:20260407T114627Z
UID:21cb1582d8c0c5d57cdce15501d27cd3cbe63209cbb741eb80c2e6c5
CATEGORIES:Conferences - Seminars
DESCRIPTION:Yubo Xie\nEDIC candidacy exam\nExam president: Prof. Martin Ra
 jman\nThesis advisor: Prof. Pearl Pu\nCo-examiner: Prof. Wulfram Gerstner\
 n\nAbstract\nIn the recent years\, the successful application of sequence 
 to sequence neural networks to statistical machine translation\, including
  the usage of attention mechanism for the aim of translation alignment\, h
 as inspired a huge amount of work in dialog generation. Most of the studie
 s are focused on how to improve the content quality of the dialogs generat
 ed\, for example\, the diverse beam search algorithm to increase the diver
 sity in the responses\, the persona-based model to handle the issue of spe
 aker consistency\, and the hierarchical recurrent encoder-decoder network 
 to model the context of multi-turn dialogs. However\, not much attention h
 as been put on the affective/emotional aspects of the dialogs. In this pro
 posal\, we introduce some existing literature on emotion handling in dialo
 g systems. Since the current work mainly deals with emotional responses in
  single-turn dialogs\, we then discuss how we could potentially extend to 
 multi-turn settings\, ideally with the capability of tracking the change o
 f emotions when the dialog progresses.\n\nBackground papers\nSequence to S
 equence Learning with Neural Networks\, by Ilya Sutskever\, et al.\nBuildi
 ng End-To-End Dialogue Systems Using Generative Hierarchical Neural Networ
 k Models\, by Iulian Vlad Serban\, et al.\nAffective Neural Response Gener
 ation\, by Nabiha Asghar\, et al.\n\n 
LOCATION:INR 212 https://plan.epfl.ch/?room=INR212
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR