Natural Language Acquisition and Grounding for Embodied Robotic Systems
- Submitting institution
-
The University of Leeds
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- UOA11-4643
- Type
- E - Conference contribution
- DOI
-
-
- Title of conference / published proceedings
- Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence
- First page
- 4349
- Volume
- 31
- Issue
- 1
- ISSN
- 2159-5399
- Open access status
- Access exception
- Month of publication
- February
- Year of publication
- 2017
- URL
-
https://ojs.aaai.org/index.php/AAAI/article/view/11161
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
B - AI (Artificial Intelligence)
- Citation count
- 5
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Grounding language is a challenging problem of increasing interest fundamentally and practically (e.g. regular CVPR/ACL workshops). First to show it is possible to concurrently and incrementally learn visual categories, language-grounding, and grammar of the language, with essentially no prior linguistic knowledge and only loose supervision from video-text pairs, producing white-box models. Demonstrated in synthetic and real worlds. Follow-ons evidencing significance: EurAI Best AI Thesis “honourable mention”(runner-up):Alomari; seven invited/keynote talks (Hogg/Cohn: VL-17, ACS-17, SpLu-18, IROS-18 workshop, R2K-18, L2A2-19, KI-20, ACS-17, L2A2-19, Cognitum-17); RoboNLP-17 Best Paper; IJCAI-17 best video; inclusion in major review https://doi.org/10.1146/annurev-control-101119-071628. Alomari now heads AI-lab at Rolls-Royce.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
