Postdoctoral Position on Robots Learning Semantic Description of Objects from Web resources - ALOOF CHIST-ERA European project

Postdoctoral Position on Robots Learning Semantic Description of Objects from Web resources - ALOOF (Autonomous Learning of the Meaning of Objects) CHIST-ERA European project
 
*Job*

We are looking for a Postdoctoral researcher with a background in Natural Language Processing and Knowledge Representation and Reasoning (in particular semantic web and linked data) to join the Inria WIMMICS team (http://wimmics.inria.fr). The context is the ALOOF (Autonomous Learning of the Meaning of Objects) CHIST-ERA European project, which goal is to enable robots and autonomous systems working with and for humans to exploit the vast amount of knowledge on the Web in order to learn about previously unseen objects involved in human activities, and to use this knowledge when acting in the real world. More precisely, the project scenario consists of an open-ended domestic setting where robots have to find objects.

Within this context, the goal of this postdoctoral position is to go beyond the current-state-of-the-art in knowledge acquisition for cognitive systems by developing and combining techniques from text mining to allow robots to engage in life-long learning from the Web. Techniques that can harvest the Web to extract symbolic knowledge about objects and their characteristics from unstructured text (relying on natural language processing and machine reading techniques), as well as available ontologies and knowledge on the Semantic Web will be developed, grounding this knowledge in visual features so that robots can recognize these objects in a real situation.

In particular, the following research tasks will be addressed:
- Building a visual object category knowledge base and a semantic object knowledge base, relying on basic ontological knowledge about objects extracted by analyzing unstructured and structured information sources on the Web following the learning by reading paradigm.
- Acquiring semantic knowledge concerning object properties from the web, to propose objects which can satisfy a particular description or role in an activity.
- Iterative cross-modal learning starting from labels of unconstrained data, to efficiently acquire knowledge about an unknown object that has been encountered in real time.

*Profile*

Mandatory requirements for applicants:
1. PhD in Computer Science;
2. Experience in NLP, Knowledge Representation and Reasoning, Semantic Web, or in a related field (Artificial Intelligence, Machine Learning...);
3. Hands-on experience of at least one programming language (e.g., Java, C++) ensure autonomy in completing the technical tasks of the project;
4. Self-motivated, goal-oriented and willing to work in an international team;
5. Fluent English is mandatory.
 
Optional:
1. Good control of scripting tools (bash, Unix/Linux tools) and of web languages;
2. Experience with automation of NLP processing chains

Project duration: 36 months
Deadline: open until filled
Working environment: the engineer will be employed at Inria Sophia Antipolis, France, in the Wimmics team
Salary: Gross Salary per month according to the level of diploma and the experience in the domain: 2500 – 2800€ / month (corresponding to 2100-2300€ net salary / month)
Contact email: Elena Cabrio: elena.cabrio@inria.fr ; Fabien Gandon: fabien.gandon@inria.fr

Received on Wednesday, 4 February 2015 13:56:16 UTC