Two PhDs in Language Games for Ontology and Computational Construction Grammar and Procedural Semantics for Multi-modal, Linguistic, Human-machine Interaction

98 views
Skip to first unread message

Frenay Benoit

unread,
May 18, 2022, 6:16:39 AM5/18/22
to Machine Learning News
Dear all,

We are looking for excellent candidates to fill two PhD positions that are currently available at the University of Namur in the field of artificial intelligence, financed by the ARIAC project. The two positions focus on the following research topics:
  • Language games for ontology alignment under partial observability
  • Computational construction grammar and procedural semantics for multi-modal, linguistic, human-machine interaction
The deadline for applying for these positions is July 1st 2022. Soon-to-be graduating master students are welcome to apply provided that they will have graduated before the start of the position. A more detailed description of the project and the application procedure is included here below.

Thank you for helping us spread the word.

Best wishes,
Katrien Beuls, Anthony Cleve and Bruno Dumas


##########################################################################################
Title
Full-time PhD researcher in artificial intelligence - Université de Namur
Language games for ontology alignment under partial observability (ARIAC project)

Project
ARIAC by DigitalWallonia4.ai is a research project funded by the Walloon Region bringing together the five French-speaking universities and four Walloon research centres with the primary objective of accelerating the development of artificial intelligence technologies in Wallonia. The project is part of the TRAIL (Trusted AI Labs) initiative, launched in September 2020, which aims at enhancing the development of artificial intelligence technologies in Wallonia.

In the context of this position, you will investigate how a population of agents can autonomously develop a new language for exchanging information. The information held by each agent might be stored in entirely different formats and according to different ontologies. Through task-based communicative interactions, the agents invent, adopt and align the building blocks of an emergent language. This language can then be used by all agents in the population to exchange information, as it forms an abstraction layer over the individual ontological organisations and storage formats.

You will carry out this PhD project both individually and in collaboration with other researchers in the field. It is conceived as a 4-year program, starting with a 2-year contract that is renewable for up to 2 additional years. You will be supervised by Prof. dr. K. Beuls and Prof. dr. A. Cleve.

Profile
  • You have obtained a master’s degree in artificial intelligence, computer science, linguistics, mathematics or a related domain. You have a solid academic track record. 
  • You are passionate about building intelligent systems that are capable of interacting with their users and environment through natural language.
  • You strive for excellence and have a scientific mindset.
  • You are a loyal team player, who can work autonomously and deliver solid scientific work.
  • You have strong communication skills and a good command of English. Knowledge of French is considered as a plus.
Additional information
For additional information please do not hesitate to contact Prof. K. Beuls or Prof. A. Cleve (see addresses below).

Important dates
Submission deadline: July 1st, 2022 (11h59 pm AoE).
Expected starting date: September 1st, 2022.

How to apply?
Applications should be sent by e-mail to secretar...@unamur.bekatrie...@unamur.be AND anthon...@unamur.be and contain the following:
  • A motivation letter describing your interest in the research topic
  • Your CV
  • A copy of your diplomas (Bachelor and Master, if available)
  • A transcript with the grades you obtained for each course taken on each university year
  • The name and e-mail address of two references to be contacted upon request

##########################################################################################
Title
Full-time PhD researcher in artificial intelligence - Université de Namur
Computational construction grammar and procedural semantics for multi-modal, linguistic, human-machine interaction (ARIAC project)

Project
ARIAC by DigitalWallonia4.ai is a research project funded by the Walloon Region bringing together the five French-speaking universities and four Walloon research centres with the primary objective of accelerating the development of artificial intelligence technologies in Wallonia. The project is part of the TRAIL (Trusted AI Labs) initiative, launched in September 2020, which aims at enhancing the development of artificial intelligence technologies in Wallonia.

Current techniques for linguistic human-machine interaction focus on either speech signals, textual data or gestures. However, communication between humans is multi-modal in nature. In particular, spoken utterances are accompanied by gestures, pointing, eye gaze, facial expressions, and prosody amongst others. These multi-modal channels are of great importance to supporting the basic function of language, i.e. exchanging information between individuals. A great challenge for human-machine interaction resides thus in the integration of multi-modal signals into a single device for comprehending and producing natural language expressions. Within the ARIAC project, we will develop a novel methodology for multi-modal language comprehension and production. This methodology will be situated within the framework of computational construction grammar, a family of linguistic theories which is well-suited to accommodate the representation and processing of linguistic knowledge that combines different sources of information. While the representation of the form of linguistic expressions will consist in abstractions over data captured through different feature channels, the representation of the meaning of linguistic expressions will be formalised using procedural semantics. In this way, the information captured in the multi-modal linguistic expressions can be executed by a machine, or a machine can produce multi-modal expressions based on procedural semantic representations. The primary application domain of this research will be systems for multi-modal human-machine interaction, i.e. systems where humans can give instructions to a machine by simultaneously speaking and gesturing.

You will carry out this PhD project both individually and in collaboration with other researchers in the field. It is conceived as a 4-year program, starting with a 2-year contract that is renewable for up to 2 additional years. You will be supervised by Prof. dr. K. Beuls and Prof. dr. B. Dumas.

Profile
  • You have obtained a master’s degree in artificial intelligence, computer science, linguistics, mathematics or a related domain. You have a solid academic track record. 
  • You are passionate about building intelligent systems that are capable of interacting with their users and environment through natural language.
  • You strive for excellence and have a scientific mindset.
  • You are a loyal team player, who can work autonomously and deliver solid scientific work.
  • You have strong communication skills and a good command of English. Knowledge of French is considered as a plus.
Additional information
For additional information please do not hesitate to contact Prof. K. Beuls or Prof. B. Dumas (see addresses below).

Important dates
Submission deadline: July 1st, 2022 (11h59 pm AoE).
Expected starting date: September 1st, 2022.

How to apply?
Applications should be sent by e-mail to secretar...@unamur.bekatrie...@unamur.be AND bruno...@unamur.be and contain the following:
  • A motivation letter describing your interest in the research topic
  • Your CV
  • A copy of your diplomas (Bachelor and Master, if available)
  • A transcript with the grades you obtained for each course taken on each university year
  • The name and e-mail address of two references to be contacted upon request


Reply all
Reply to author
Forward
0 new messages