Two members of the Interactive Robotics Group will be presenting their work at NeurIPS this week. Ankit Shah will be presenting his main track paper on Bayesian Inference of Temporal Task Specifications at poster #107 at 5.00 PM on Tuesday in 210 AB at the Palais des Congres, Montreal. This paper focuses on inferring task specifications as LTL formulas thus providing a clear demarcation between acceptable and unacceptable task performance. You can check out the spotlight video below.
Claudia Pérez D'Arpino will be presenting at the following NeurIPS Workshops:
- NeurIPS Workshop on Modeling the Physical World: Perception, Learning, and Control on Friday 12/7 in room 517 C
Poster: Observational and self-learning of multi-step robotic manipulation with unknown physical properties
- NeurIPS LXAI Research Workshop on Saturday 12/8 in room 520
Talk at 5:30pm and Poster: Learning how to plan for multi-step manipulation in collaborative robotics
Spotlight video for Bayesian Inference of Temporal Task Specifications by Ankit Shah: