Imagining Grounded Conceptual Representations from Perceptual Information in Situated Guessing Games

Alessandro Suglia, Antonio Vergari, Ioannis Konstas, Yonatan Bisk, Emanuele Bastianelli, Andrea Vanzo, Oliver Lemon

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Downloads (Pure)

Abstract

In visual guessing games, a Guesser has to identify a target object in a scene by asking questions to an Oracle. An effective strategy for the players is to learn conceptual representations of objects that are both discriminative and expressive enough to ask questions and guess correctly. However, as shown by Suglia et al. (2020), existing models fail to learn truly multi-modal representations, relying instead on gold category labels for objects in the scene both at training and inference time. This provides an unnatural performance advantage when categories at inference time match those at training time, and it causes models to fail in more realistic “zero-shot” scenarios where out-of-domain object categories are involved. To overcome this issue, we introduce a novel “imagination” module based on Regularized Auto-Encoders, that learns context-aware and category-aware latent embeddings without relying on category labels at inference time. Our imagination module outperforms state-of-the-art competitors by 8.26% gameplay accuracy in the CompGuessWhat?! zero-shot scenario (Suglia et al., 2020), and it improves the Oracle and Guesser accuracy by 2.08% and 12.86% in the GuessWhat?! benchmark, when no gold categories are available at inference time. The imagination module also boosts reasoning about object properties and attributes.
Original languageEnglish
Title of host publicationProceedings of the 28th International Conference on Computational Linguistics
PublisherAssociation for Computational Linguistics
Pages1090–1102
Number of pages13
ISBN (Electronic)9781952148279
DOIs
Publication statusPublished - Dec 2020

Keywords

  • cs.CL
  • cs.CV
  • cs.LG

Fingerprint

Dive into the research topics of 'Imagining Grounded Conceptual Representations from Perceptual Information in Situated Guessing Games'. Together they form a unique fingerprint.

Cite this