BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Building Compassionate Chatbots Using Neural Network Models
DTSTART:20190709T153000
DTEND:20190709T173000
DTSTAMP:20260410T152247Z
UID:3312879452bab224fa06d47875f3abe26b603dc81030b78b632f5822
CATEGORIES:Conferences - Seminars
DESCRIPTION:Anuradha Welivita\nEDIC candidacy exam\nExam president: Dr. Ma
 rtin Rajman\nThesis advisor: Dr. Pearl Pu Faltings\nCo-examiner: Prof. Rob
 ert West\n\nAbstract\nOpen-domain conversational agents or chatbots have i
 ncreasingly become popular in the natural language processing community. I
 n recent years\, extensive research has been conducted on building open-do
 main conversational models using neural networks. Even though these models
  can process natural language at a lexico-syntactic level\, they often fai
 l to identify subtle variations of emotion or affect in human conversation
 s and respond in an emotionally appropriate manner. Our research goal is t
 o identify novel approaches in building compassionate neural chatbots that
  can effectively identify and process emotional content present in user ut
 terances and generate responses that mimic human social and emotional inte
 lligence. We discuss three existing works related to neural conversational
  agents and emotion analysis and how they relate to our work. We first exa
 mine a basic sequence-to-sequence neural network model for generating huma
 n-like responses in a conversation. We then look at how this model can be 
 augmented with affect information based on the traditional Valence-Arousal
 -Dissonance (VAD) affective notation\, affect-based objective functions an
 d affectively diverse decoding strategies to generate responses that are r
 ich in emotion. Finally\, we discuss a novel approach to learn richer repr
 esentations of human emotions\, using emojis used in social media.\n\nBack
 ground papers\nA neural conversational model\, by  O. Vinyals\, and Q.
  Le. In Proceedings of the 31st International Conference on Machine Lea
 rning\, volume 37\, 2015.\nAffective neural response generation\, by N. A
 sghar\, P. Poupart\, J. Hoey\, X. Jiang\, and L. Mou. In European C
 onference on Information Retrieval\, pages 154-166\, Springer\, Cham\, 2
 018.\nUsing millions of emoji occurrences to learn any-domain representati
 ons for detecting sentiment\, emotion and sarcasm\, by  B. Felbo\, A. 
 Mislove\, A. Søgaard\, I. Rahwan\, and S. Lehmann. In Proceedings 
 of the 2017 Conference on Empirical Methods in Natural Language Processing
 \, pages 1615–1625\, 2017.
LOCATION:INR 212 https://plan.epfl.ch/?room=INR212
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR