Elamur Research⌕
☰
home / papers / 2015 / Learning in the Rational Speech Acts Model

Learning in the Rational Speech Acts Model

Paper · 2015 · Will Monroe and Christopher Potts

Summary

The Rational Speech Acts (RSA) model describes pragmatics: the speaker and the listener recursively model each other's intentions and thereby enrich the literal meaning of utterances. However, classical RSA has two shortcomings. The semantic lexicon has to be specified by hand, and the speakers in the model turn out to be more rational than real people. The authors propose learned RSA: the model becomes a statistical classifier in which the intermediate RSA agents act as hidden layers with a nonlinear activation. The model is built on a log-linear literal speaker S0. The model learns the lexicon from data and optimizes its parameters through the likelihood of the pragmatic speaker S1. Thanks to arbitrary features, preferences known from natural language generation can be added to the model, for example the tendency to mention an object's color or type. The model was evaluated on attribute selection for referring expressions in the TUNA corpus. Learned RSA outperforms both pure RSA and an ordinary classifier, most noticeably on the more difficult people subdomain. The work shows that pragmatic reasoning can be learned from data. This opens the way to applying RSA in NLP.

Key points

  • TUNA corpus, singular trials in the furniture and people domains: the target is shown among seven distractors, and the task is to reproduce the multiset of attributes in the description. The metrics are accuracy and multiset Dice, evaluation uses five-fold cross-validation, and significance was tested with the Wilcoxon test.
  • Two types of features: basic indicators for "object attribute × message attribute" pairs and generation features (attribute type, pairwise type matches, message size). Only together do they yield the best result.
  • The best model, Learned S1 with basic and generation features, achieves 25.3% accuracy and a Dice of .777 overall. Pure RSA s1 achieves only 2.2% and .386.
  • On people, the S1 model reaches 22.5% accuracy and a Dice of .764. This is slightly above the best dev result of the 2008 Challenge (.762), although the authors stress that the comparison is informal.
  • Pure RSA rarely mentions color and head nouns, while the S0 models underuse informative attributes such as beard and hair. Learned S1 with generation features corrects both systematic errors.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Stanford UniversityStanford · United States434 works · 314,419 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

Cite

Cite as Will Monroe and Christopher Potts. 2015. Learning in the Rational Speech Acts Model. arXiv:1510.06807.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.1510.06807 · arXiv 1510.06807

@misc{monroe2015learning,
  title = {{Learning in the Rational Speech Acts Model}},
  author = {Will Monroe and Christopher Potts},
  year = {2015},
  doi = {10.48550/arxiv.1510.06807},
  eprint = {1510.06807},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.1510.06807}
}

References

36 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Peter Baumann, Brady Clark, and Stefan Kaufmann. Overspecification and the cost of pragmatic reasoning about referring expressions. In Proceedings of the 36th Annual Meeting of the Cognitive Science Society. Cognitive Science Society, 2014. find
  2. Leon Bergen, Roger Levy, and Noah D. Goodman. Pragmatic reasoning through semantic inference. Ms., MIT, UCSD, and Stanford, 2014. find
  3. Léon Bottou. Large-scale machine learning with stochastic gradient descent. In Yves Lechevallier and Gilbert Saporta, editors, Proceedings of the 19th International Conference on Computational Statistics, pages 177–186. Springer, Berlin, 2010. ↗
  4. Colin F. Camerer, Teck-Hua Ho, and Juin-Kuan Chong. A cognitive hierarchy model of games. The Quarterly Journal of Economics, 119(3):861–898, 2004. ↗
  5. Stanley F. Chen and Ronald Rosenfeld. A Gaussian prior for smoothing maximum entropy models. Technical Report CMU-CS-99-108, Carnegie Mellon University, 1999. ↗
  6. Robert Dale and Ehud Reiter. Computational interpretations of the Gricean maxims in the generation of referring expressions. Cognitive Science, 19(2):233–263, 1995. ↗
  7. Judith Degen and Michael Franke. Optimal reasoning about referential expressions. In Sarah Brown-Schmidt, Jonathan Ginzburg, and Staffan Larsson, editors, Proceedings of the 16th Workshop on the Semantics and Pragmatics of Dialogue, pages 2–11, 2012. find
  8. Pedro Domingos. A few useful things to know about machine learning. Communications of ACM, 55(10):78–87, 2012. ↗
  9. John Duchi, Elad Hazan, and Yoram Singer. Adaptive subgradient methods for online learning and stochastic optimization. Journal of Machine Learning Research, pages 2121–2159, 2011. find
  10. Paul E. Engelhardt, Karl G.D. Bailey, and Fernanda Ferreira. Do speakers and listeners observe the Gricean maxim of quantity? Journal of Memory and Language, 54(4):554–573, 2006. ↗
  11. Michael C. Frank and Noah D. Goodman. Predicting pragmatic reasoning in language games. Science, 336(6084):998, 2012. ↗
  12. Michael C. Frank and Noah D. Goodman. Inferring word meanings by assuming that speakers are informative. Cognitive Psychology, 75(1):80–96, 2014. ↗
  13. Michael Franke. Signal to Act: Game Theory in Pragmatics. ILLC Dissertation Series. Institute for Logic, Language and Computation, University of Amsterdam, 2009. find
  14. Albert Gatt, Anja Belz, and Eric Kow. The TUNA Challenge 2008: Overview and evaluation results. In Proceedings of the Fifth International Natural Language Generation Conference, INLG ’08, pages 198–206. Association for Computational Linguistics, 2008. find
  15. Albert Gatt, Ielka van der Sluis, and Kees van Deemter. Evaluating algorithms for the generation of referring expressions using a balanced corpus. In Proceedings of the Eleventh European Workshop on Natural Language Generation, pages 49–56. DFKI GmbH, 2007. find
  16. Albert Gatt, Roger P. G. van Gompel, Kees van Deemter, and Emiel Krahmer. Are we Bayesian referring expression generators? In Proceedings of the Thirty-Fifth Annual Conference of the Cognitive Science Society, 2013. Monroe and Potts Learning in the Rational Speech Acts Model find
  17. Dave Golland, Percy Liang, and Dan Klein. A game-theoretic approach to generating spatial descriptions. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 410–419. Association for Computational Linguistics, 2010. find
  18. H. Paul Grice. Logic and conversation. In Peter Cole and Jerry Morgan, editors, Syntax and Semantics, volume 3: Speech Acts, pages 43–58. Academic Press, 1975. ↗
  19. Trevor Hastie, Robert Tibshirani, and Jerome Friedman. The Elements of Statistical Learning. Springer, Berlin, 2 edition, 2009. ↗
  20. Gerhard Jäger. Game theory in semantics and pragmatics. In Claudia Maienborn, Klaus von Heusinger, and Paul Portner, editors, Semantics: An International Handbook of Natural Language Meaning, volume 3, pages 2487–2425. Mouton de Gruyter, 2012. find
  21. Gerhard Jäger. Rationalizable signaling. Erkenntnis, 79(4):673–706, 2014. find
  22. Justine T. Kao, Leon Bergen, and Noah D. Goodman. Formalizing the pragmatics of metaphor understanding. In Proceedings of the 36th Annual Meeting of the Cognitive Science Society, pages 719–724. Cognitive Science Society, 2014. find
  23. Justine T. Kao, Jean Y. Wu, Leon Bergen, and Noah D. Goodman. Nonliteral understanding of number words. Proceedings of the National Academy of Sciences, 111(33):12002–12007, 2014. ↗
  24. Willem JM Levelt. Speaking: From Intention to Articulation, volume 1. MIT press, 1993. find
  25. David Lewis. Convention. Harvard University Press, 1969. Reprinted 2002 by Blackwell. find
  26. Percy Liang and Christopher Potts. Bringing machine learning and compositional semantics together. Annual Review of Linguistics, 1(1):355–376, 2015. ↗
  27. Peter McCullagh and John A. Nelder. Generalized Linear Models. Chapman and Hall, London, 1989. ↗
  28. Brian McMahan and Matthew Stone. A Bayesian model of grounded color semantics. Transactions of the Association for Computational Linguistics, 3:103–115, 2015. ↗
  29. Thomas Pechmann. Incremental speech production and referential overspecification. Linguistics, 27(1):89–110, 1989. ↗
  30. Christopher Potts, Daniel Lassiter, Roger Levy, and Michael C. Frank. Embedded implicatures as pragmatic inferences under compositional lexical uncertainty. To appear in Journal of Semantics, 2015. ↗
  31. Seymour Rosenberg and Bertram D. Cohen. Speakers’ and listeners’ processes in a word communication task. Science, 145:1201–1203, 1964. ↗
  32. David E. Rumelhart, Geoffrey E. Hinton, and Ronald J. Williams. Learning internal representations by error propagation. In David E. Rumelhart and James L. McClelland, editors, Parallel Distributed Processing: Explorations in the Microstructure of Cognition, volume 1: Foundations, pages 318–362. MIT Press, Cambridge, MA, 1986. ↗
  33. Alex Stiller, Noah D. Goodman, and Michael C. Frank. Ad-hoc scalar implicature in adults and children. In Laura Carlson, Christoph Hoelscher, and Thomas F. Shipley, editors, Proceedings of the 33rd Annual Meeting of the Cognitive Science Society, pages 2134–2139. Cognitive Science Society, 2011. find
  34. Kees van Deemter, Ielka van der Sluis, and Albert Gatt. Building a semantically transparent corpus for the generation of referring expressions. In Proceedings of the Fourth International Natural Language Generation Conference, pages 130–132. Association for Computational Linguistics, 2006. find
  35. Adam Vogel, Max Bodoia, Christopher Potts, and Dan Jurafsky. Emergence of Gricean maxims from multi-agent decision theory. In Human Language Technologies: The 2013 Annual Conference of the North American Chapter of the Association for Computational Linguistics, pages 1072–1081. Association for Computational Linguistics, 2013.
  36. Adam Vogel, Andrés Gómez Emilsson, Michael C. Frank, Dan Jurafsky, and Christopher Potts. Learning to reason pragmatically with cognitive limitations. In Proceedings of the 36th Annual Meeting of the Cognitive Science Society, pages 3055–3060. Cognitive Science Society, 2014. Monroe and Potts find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/monroe-2015-learning-rational-speech-acts-model/