Learning to Request Guidance in Emergent Communication
@article{Kolb2019LearningTR, title={Learning to Request Guidance in Emergent Communication}, author={B. Kolb and Leon Lang and H. Bartsch and Arwin Gansekoele and Raymond Koopmanschap and Leonardo Romor and David Speck and M. Mul and Elia Bruni}, journal={ArXiv}, year={2019}, volume={abs/1912.05525} }
Previous research into agent communication has shown that a pre-trained guide can speed up the learning process of an imitation learning agent. The guide achieves this by providing the agent with discrete messages in an emerged language about how to solve the task. We extend this one-directional communication by a one-bit communication channel from the learner back to the guide: It is able to ask the guide for help, and we limit the guidance by penalizing the learner for these requests. During… CONTINUE READING
Figures and Topics from this paper
References
SHOWING 1-10 OF 28 REFERENCES
Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols
- Computer Science, Mathematics
- NIPS
- 2017
- 126
- PDF
Vision-Based Navigation With Language-Based Assistance via Imitation Learning With Indirect Intervention
- Computer Science, Mathematics
- 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- 2019
- 27
- PDF
Scalable agent alignment via reward modeling: a research direction
- Computer Science, Mathematics
- ArXiv
- 2018
- 59
- PDF
Multi-Agent Cooperation and the Emergence of (Natural) Language
- Computer Science
- ICLR
- 2017
- 219
- Highly Influential
- PDF
Delegative Reinforcement Learning: learning to avoid traps with a little help
- Mathematics, Computer Science
- ArXiv
- 2019
- 3
- PDF
Mapping Instructions and Visual Observations to Actions with Reinforcement Learning
- Computer Science
- EMNLP
- 2017
- 125
- PDF