Generalizing Emergent Communication.
@article{Unger2020GeneralizingEC, title={Generalizing Emergent Communication.}, author={Thomas A. Unger and Elia Bruni}, journal={arXiv: Artificial Intelligence}, year={2020} }
We converted the recently developed BabyAI grid world platform to a sender/receiver setup in order to test the hypothesis that established deep reinforcement learning techniques are sufficient to incentivize the emergence of a grounded discrete communication protocol between generalized agents. This is in contrast to previous experiments that employed straight-through estimation or specialized inductive biases. Our results show that these can indeed be avoided, by instead providing proper… CONTINUE READING
References
SHOWING 1-10 OF 20 REFERENCES
Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols
- Computer Science, Mathematics
- NIPS
- 2017
- 126
- PDF
Emergence of Linguistic Communication from Referential Games with Symbolic and Pixel Input
- Computer Science
- ICLR
- 2018
- 94
- PDF
Scalable agent alignment via reward modeling: a research direction
- Computer Science, Mathematics
- ArXiv
- 2018
- 59
- PDF
Intrinsic Social Motivation via Causal Influence in Multi-Agent RL
- Computer Science, Mathematics
- ArXiv
- 2018
- 39
- PDF
Emergence of Grounded Compositional Language in Multi-Agent Populations
- Computer Science
- AAAI
- 2018
- 300
- PDF