1 |
Communicating artificial neural networks develop efficient color-naming systems
|
|
|
|
In: ISSN: 0027-8424 ; EISSN: 1091-6490 ; Proceedings of the National Academy of Sciences of the United States of America ; https://hal.inria.fr/hal-03329084 ; Proceedings of the National Academy of Sciences of the United States of America , National Academy of Sciences, 2021, 118 (12), ⟨10.1073/pnas.2016569118⟩ (2021)
|
|
BASE
|
|
Show details
|
|
2 |
On the proper role of linguistically-oriented deep net analysis in linguistic theorizing ...
|
|
|
|
BASE
|
|
Show details
|
|
4 |
Communicating artificial neural networks develop efficient color-naming systems
|
|
|
|
In: Proc Natl Acad Sci U S A (2021)
|
|
BASE
|
|
Show details
|
|
5 |
Compositionality and Generalization in Emergent Languages
|
|
|
|
In: ACL 2020 - 8th annual meeting of the Association for Computational Linguistics ; https://hal.archives-ouvertes.fr/hal-02959466 ; ACL 2020 - 8th annual meeting of the Association for Computational Linguistics, Jul 2020, Seattle / Virtual, United States (2020)
|
|
Abstract:
International audience ; Natural language allows us to refer to novel composite concepts by combining expressions denoting their parts according to systematic rules, a property known as compositional-ity. In this paper, we study whether the language emerging in deep multi-agent simulations possesses a similar ability to refer to novel primitive combinations, and whether it accomplishes this feat by strategies akin to human-language compositionality. Equipped with new ways to measure compositionality in emergent languages inspired by disentan-glement in representation learning, we establish three main results. First, given sufficiently large input spaces, the emergent language will naturally develop the ability to refer to novel composite concepts. Second, there is no correlation between the degree of compositional-ity of an emergent language and its ability to generalize. Third, while compositionality is not necessary for generalization, it provides an advantage in terms of language transmission: The more compositional a language is, the more easily it will be picked up by new learners, even when the latter differ in architecture from the original agents. We conclude that compositionality does not arise from simple generalization pressure, but if an emergent language does chance upon it, it will be more likely to survive and thrive.
|
|
Keyword:
[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]; [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL]; [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG]
|
|
URL: https://hal.archives-ouvertes.fr/hal-02959466/document https://hal.archives-ouvertes.fr/hal-02959466/file/2004.09124.pdf https://hal.archives-ouvertes.fr/hal-02959466
|
|
BASE
|
|
Hide details
|
|
7 |
Mechanisms for Handling Nested Dependencies in Neural-Network Language Models and Humans ...
|
|
|
|
BASE
|
|
Show details
|
|
8 |
A Benchmark for Systematic Generalization in Grounded Language Understanding ...
|
|
|
|
BASE
|
|
Show details
|
|
9 |
Linguistic generalization and compositionality in modern artificial neural networks
|
|
|
|
In: Philos Trans R Soc Lond B Biol Sci (2020)
|
|
BASE
|
|
Show details
|
|
10 |
Anti-efficient encoding in emergent communication
|
|
|
|
In: https://hal.archives-ouvertes.fr/hal-02274205 ; 2019 (2019)
|
|
BASE
|
|
Show details
|
|
11 |
Word-order biases in deep-agent emergent communication
|
|
|
|
In: ACL 2019 - 57th Annual Meeting of the Association for Computational Linguistics ; https://hal.archives-ouvertes.fr/hal-02274157 ; ACL 2019 - 57th Annual Meeting of the Association for Computational Linguistics, Jul 2019, Florence, Italy (2019)
|
|
BASE
|
|
Show details
|
|
12 |
EGG: a toolkit for research on Emergence of lanGuage in Games
|
|
|
|
In: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations ; https://hal.archives-ouvertes.fr/hal-02274229 ; Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations, Nov 2019, Hong Kong, China. ⟨10.18653/v1/D19-3010⟩ (2019)
|
|
BASE
|
|
Show details
|
|
13 |
On the Distribution of Deep Clausal Embeddings: A Large Cross-linguistic Study ...
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Focus on What's Informative and Ignore What's not: Communication Strategies in a Referential Game ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
Tabula nearly rasa: Probing the Linguistic Knowledge of Character-Level Neural Language Models Trained on Unsegmented Text ...
|
|
|
|
BASE
|
|
Show details
|
|
16 |
Colorless green recurrent networks dream hierarchically
|
|
|
|
In: Proceedings of the Society for Computation in Linguistics (2019)
|
|
BASE
|
|
Show details
|
|
17 |
What you can cram into a single \$&!#* vector: Probing sentence embeddings for linguistic properties
|
|
|
|
In: ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics ; https://hal.archives-ouvertes.fr/hal-01898412 ; ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Jul 2018, Melbourne, Australia. pp.2126-2136 (2018)
|
|
BASE
|
|
Show details
|
|
18 |
How agents see things: On visual representations in an emergent language game ...
|
|
|
|
BASE
|
|
Show details
|
|
19 |
What you can cram into a single vector: Probing sentence embeddings for linguistic properties ...
|
|
|
|
BASE
|
|
Show details
|
|
20 |
Colorless green recurrent networks dream hierarchically ...
|
|
|
|
BASE
|
|
Show details
|
|
|
|