Overview

PhD position in Computational models of Disfluencies in social interactions between a pedagogical agent and a student

Place of work : Telecom ParisTech [1] 46 rue Barrault 75013 Paris ? France Paris until 2019, and then Palaiseau (Paris outskirt)

Starting date: March 2018

Duration of the PhD : 36 months

*Position description*

The PhD student will take part in ANIMATAS which is a H2020 Marie Sklodowska Curie European Training Network funded by Horizon 2020 (the European Union?s Framework Programme for Research and Innovation).
ANIMATAS focuses on the following objectives:
1) Exploration of fundamental questions relating to the interconnections between robots and virtual characters? appearance, behaviours and perception by people
2) Development of new social learning mechanisms that can deal with different types of human intervention and allow robots and virtual characters to learn in an unconstrained manner
3) Development of new approaches for robots and virtual characters? personalised adaptation to human users in unstructured and dynamically evolving social interactions
Within this project, the PhD student will focus on the development of a computational model of student?s disfluencies during his/her social interactions with a pedagogical agent (e.g. a robot in a school), the other students or a teacher. Disfluencies (fillers, repetitions, auto-corrections, etc.) are spontaneous speech phenomena that can be linked to various factors such as speaker?s emotions, or speaker?s feeling of knowing. Studies have shown they can be markers of student?s stress, feeling of incompetence. Despite the potential of disfluencies to provide information on user?s emotion and mental state, the consideration of disfluencies in human-agent context has been so far rarely explored.
The role of the PhD will consist in analyzing the different types of disfluencies occurring in the student?s speech and in proposing a computational model (both formal models and machine learning models could be investigated according to Phd student background) of disfluencies that will allow the robot to adapt its answer.
The work will include:
a pluridisciplinary review of literature ranging from sociolinguistic corpus studies to computational linguistics and human-machine interaction on: i) disfluency studies; ii) teaching strategies
the study of acoustic and linguistic realizations of the different types of disfluencies and their link with the accompanying gestures
a focus on linguistic and acoustic features of disfluencies that can give information about the student?s feeling of competence and emotional state.
the development of a computational model (from formal models to machine learning) that will formalize the teaching and social interaction strategies according to the detected disfluencies (i.e. what the robot can do and when it is relevant to trigger the strategy in order to help the student in his/her learning phase).

The PhD will join the Social Computing topic [2] in the S2a group [3] at Telecom-ParisTech in collaboration with ISIR [6].

Selected references for this position :
Cafaro, A., Glas, N. and Pelachaud, C., 2016, May. The effects of interrupting behavior on interpersonal attitude and engagement in dyadic interactions. In Proceedings of the 2016 International Conference on Autonomous Agents & Multiagent Systems (pp. 911-920). International Foundation for Autonomous Agents and Multiagent Systems.
S. Campano, C. Clavel, C. Pelachaud, « I like this painting too »: when an ECA shares appreciations to engage users, in AAMAS 2015, Istanbul, Turkey.
Clavel, C., Cafaro, A., Campano, S., & Pelachaud, C. (2016). Fostering user engagement in face-to-face human-agent interactions: a survey. In Toward Robotic Socially Believable Behaving Systems-Volume II (pp. 93-120). Springer International Publishing.
C. Dutrey, C. Clavel, S. Rosset, I. Vasilescu, and M. Adda-Decker, ?A crf-based approach to automatic disfluency detection in a french call-centre corpus,? in Interspeech, 2014.
Vasilescu, I., Rosset, S. and Adda-Decker, M., 2010. On the Functions of the Vocalic Hesitation euh in Interactive Man-machine Question Answering Dialogs in French. In Proceedings of DiSS- LPSS Joint Workshop 2010/The 5th Workshop on Disfluency in Spontaneous Speech – The 2nd International Symposium on Linguistic Patterns in Spontaneous Speech, Tokyo, Japan.

Company:

Telecom ParisTech

Qualifications:

Criteria for Eligibility:
? Eligible candidates must be in the first 4 years of their research career, and have not yet been awarded a doctorate degree.
? Eligible candidates must not have resided or carried out their main activity in the country of his host institution for more than 12 months in the 3 years immediately prior to his/her recruitment under the project (compulsory national service and/or short stays such as holidays are not taken into account).
? Eligible candidates can be of any nationality.

As a minimum requirement, the successful candidate will have:

? A master degree or equivalent in one or more of the following areas: human-agent interaction, machine
learning, computational linguistics, affective computing
? Excellent programming skills (preferably in Python)

Language requirements:

? Good command of English

Educational level:

Master Degree

Level of experience (years):

Mid Career (2+ years of experience)

Tagged as: , , ,

You can apply to this job and others using your online resume. Click the link below to submit your online resume and email your application to this employer.

About Télécom ParisTech

Télécom ParisTech (also known as ENST or École nationale supérieure des télécommunications) is one of the most prestigious and selective engineering school in France and one of the finest institutions in the field of Telecommunications. Located in Paris, it is also a member of ParisTech and Institut Mines-Télécom.