Opportunistic Active Learning for Grounding Natural Language Descriptions
By Jesse Thomason, Aishwarya Padmakumar, Jivko Sinapov, Justin W. Hart, Peter Stone, Raymond J. Mooney. 2017
49 citations recorded by the source corpus.
Subfields: Reinforcement Learning
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Reinforcement Learning
- Yonatan Bisk
- Cynthia Matuszek
- Jacob Andreas
- Jivko Sinapov
- Nakul Gopalan
- Yoshua Bengio
- Nicholas A. Roy
- Jesse Thomason
Citations in the corpus
- Experience Grounds Language (2020, background): “Queries can be resolved via dialog-based exploration with a human interlocutor (Liu and Chai, 2015), even as new object properties like texture, weight, and sound become available (Thomason et al., 2017).”
- Robots That Use Language (2020, background): “A bottle consists of a bottle and a cap, and a person referencing “the bottle” may mean both, or they may say, “Grab the bottle and then turn the cap,”referring”
- Jointly Improving Parsing and Perception for Natural Language Commands through Human-Robot Dialog (2020, methodology): “Prior work established that an agent asking questions about both on-topic (used in the current conversation) and off-topic (irrelevant to the current user’s needs) predicates outperformed an agent that only asked about predicates in the current human description [39].”
- Weakly-Supervised Video Object Grounding from Text by Loss Weighting and Object Interaction (2018, background): “This task is often formulated as a visual-semantic alignment problem [7] and has broad applications including retrieval [7, 8], description generation [20, 25], and human-robot interaction [1, 22].”
- Learning Interpretable Spatial Operations in a Rich 3D Blocks World (2017, background): “2011), as acquiring a large corpus of natural language grounded with a real robot is prohibitively expensive (Misra et al. 2014; Thomason et al. 2017).”
- Grounding Robot Plans from Natural Language Instructions with Incomplete World Knowledge (2018, background): “Other efforts have focused on acquiring abstract concepts by interacting with the user [4, 27, 28, 29, 30].”