Matches in SemOpenAlex for { <https://semopenalex.org/work/W3208148973> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W3208148973 abstract "While multi-agent reinforcement learning has been used as an effective means to study emergent communication between agents, existing work has focused almost exclusively on communication with discrete symbols. Human communication often takes place (and emerged) over a continuous acoustic channel; human infants acquire language in large part through continuous signalling with their caregivers. We therefore ask: Are we able to observe emergent language between agents with a continuous communication channel trained through reinforcement learning? And if so, what is the impact of channel characteristics on the emerging language? We propose an environment and training methodology to serve as a means to carry out an initial exploration of these questions. We use a simple messaging environment where a speaker agent needs to convey a concept to a listener. The Speaker is equipped with a vocoder that maps symbols to a continuous waveform, this is passed over a lossy continuous channel, and the Listener needs to map the continuous signal to the concept. Using deep Q-learning, we show that basic compositionality emerges in the learned language representations. We find that noise is essential in the communication channel when conveying unseen concept combinations. And we show that we can ground the emergent communication by introducing a caregiver predisposed to hearing or speaking English. Finally, we describe how our platform serves as a starting point for future work that uses a combination of deep reinforcement learning and multi-agent systems to study our questions of continuous signalling in language learning and emergence." @default.
- W3208148973 created "2021-11-08" @default.
- W3208148973 creator A5016518233 @default.
- W3208148973 creator A5017043594 @default.
- W3208148973 creator A5040305929 @default.
- W3208148973 creator A5078537643 @default.
- W3208148973 creator A5091413322 @default.
- W3208148973 date "2021-11-04" @default.
- W3208148973 modified "2023-09-27" @default.
- W3208148973 title "Towards Learning to Speak and Hear Through Multi-Agent Communication over a Continuous Acoustic Channel." @default.
- W3208148973 cites W1640915152 @default.
- W3208148973 cites W1757796397 @default.
- W3208148973 cites W1976274322 @default.
- W3208148973 cites W2020944885 @default.
- W3208148973 cites W2110045175 @default.
- W3208148973 cites W2121863487 @default.
- W3208148973 cites W2148154194 @default.
- W3208148973 cites W2150961342 @default.
- W3208148973 cites W2166188321 @default.
- W3208148973 cites W2173772347 @default.
- W3208148973 cites W2193413348 @default.
- W3208148973 cites W2346794446 @default.
- W3208148973 cites W2501649918 @default.
- W3208148973 cites W2546861836 @default.
- W3208148973 cites W2963658727 @default.
- W3208148973 cites W2963681240 @default.
- W3208148973 cites W2963881016 @default.
- W3208148973 cites W3002559132 @default.
- W3208148973 cites W3014436489 @default.
- W3208148973 cites W3016208931 @default.
- W3208148973 cites W3033076904 @default.
- W3208148973 cites W3034438474 @default.
- W3208148973 cites W3034622111 @default.
- W3208148973 cites W3099434687 @default.
- W3208148973 cites W3112984878 @default.
- W3208148973 hasPublicationYear "2021" @default.
- W3208148973 type Work @default.
- W3208148973 sameAs 3208148973 @default.
- W3208148973 citedByCount "0" @default.
- W3208148973 crossrefType "posted-content" @default.
- W3208148973 hasAuthorship W3208148973A5016518233 @default.
- W3208148973 hasAuthorship W3208148973A5017043594 @default.
- W3208148973 hasAuthorship W3208148973A5040305929 @default.
- W3208148973 hasAuthorship W3208148973A5078537643 @default.
- W3208148973 hasAuthorship W3208148973A5091413322 @default.
- W3208148973 hasConcept C101765175 @default.
- W3208148973 hasConcept C107457646 @default.
- W3208148973 hasConcept C127162648 @default.
- W3208148973 hasConcept C154945302 @default.
- W3208148973 hasConcept C41008148 @default.
- W3208148973 hasConcept C76155785 @default.
- W3208148973 hasConcept C97541855 @default.
- W3208148973 hasConceptScore W3208148973C101765175 @default.
- W3208148973 hasConceptScore W3208148973C107457646 @default.
- W3208148973 hasConceptScore W3208148973C127162648 @default.
- W3208148973 hasConceptScore W3208148973C154945302 @default.
- W3208148973 hasConceptScore W3208148973C41008148 @default.
- W3208148973 hasConceptScore W3208148973C76155785 @default.
- W3208148973 hasConceptScore W3208148973C97541855 @default.
- W3208148973 hasLocation W32081489731 @default.
- W3208148973 hasOpenAccess W3208148973 @default.
- W3208148973 hasPrimaryLocation W32081489731 @default.
- W3208148973 hasRelatedWork W2074456508 @default.
- W3208148973 hasRelatedWork W2168327578 @default.
- W3208148973 hasRelatedWork W2284650039 @default.
- W3208148973 hasRelatedWork W2342671102 @default.
- W3208148973 hasRelatedWork W2564324149 @default.
- W3208148973 hasRelatedWork W2627585944 @default.
- W3208148973 hasRelatedWork W2887012691 @default.
- W3208148973 hasRelatedWork W2888923036 @default.
- W3208148973 hasRelatedWork W2911810244 @default.
- W3208148973 hasRelatedWork W2963155490 @default.
- W3208148973 hasRelatedWork W2989265781 @default.
- W3208148973 hasRelatedWork W3008022385 @default.
- W3208148973 hasRelatedWork W3016099117 @default.
- W3208148973 hasRelatedWork W3037976278 @default.
- W3208148973 hasRelatedWork W3080185952 @default.
- W3208148973 hasRelatedWork W3103247365 @default.
- W3208148973 hasRelatedWork W3163795532 @default.
- W3208148973 hasRelatedWork W3185935397 @default.
- W3208148973 hasRelatedWork W3206039297 @default.
- W3208148973 hasRelatedWork W1936520471 @default.
- W3208148973 isParatext "false" @default.
- W3208148973 isRetracted "false" @default.
- W3208148973 magId "3208148973" @default.
- W3208148973 workType "article" @default.