Supervised Seeded Iterated Learning for Interactive Language Learning
Yuchen Lu, Soumye Singhal, Florian Strub, Olivier Pietquin, Aaron Courville
Abstract
Language drift has been one of the major obstacles to train language models through interaction. When word-based conversational agents are trained towards completing a task, they tend to invent their language rather than leveraging natural language. In recent literature, two general methods partially counter this phenomenon: Supervised Selfplay (S2P) and Seeded Iterated Learning (SIL). While S2P jointly trains interactive and supervised losses to counter the drift, SIL changes the training dynamics to prevent language drift from occurring. In this paper, we first highlight their respective weaknesses, i.e., late-stage training collapses and higher negative likelihood when evaluated on human corpus. Given these observations, we introduce Supervised Seeded Iterated Learning (SSIL) to combine both methods to minimize their respective weaknesses. We then show the effectiveness of in the language-drift translation game.- Anthology ID:
- 2020.emnlp-main.325
- Volume:
- Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
- Month:
- November
- Year:
- 2020
- Address:
- Online
- Editors:
- Bonnie Webber, Trevor Cohn, Yulan He, Yang Liu
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 3962–3970
- Language:
- URL:
- https://aclanthology.org/2020.emnlp-main.325
- DOI:
- 10.18653/v1/2020.emnlp-main.325
- Cite (ACL):
- Yuchen Lu, Soumye Singhal, Florian Strub, Olivier Pietquin, and Aaron Courville. 2020. Supervised Seeded Iterated Learning for Interactive Language Learning. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 3962–3970, Online. Association for Computational Linguistics.
- Cite (Informal):
- Supervised Seeded Iterated Learning for Interactive Language Learning (Lu et al., EMNLP 2020)
- PDF:
- https://preview.aclanthology.org/corrections-2024-05/2020.emnlp-main.325.pdf
- Data
- Multi30K