Fig 1.
In this example, the speaker has to communicate the concept of a blue circle and the listener has to identify all blue circles. The loss function rewards shorter messages from the speaker (efficiency) and correct selections from the listener (informativeness). This figure was created by the first author.
Fig 2.
Game scenarios in Experiment 1.
A: Context-unaware - The speaker agent only observes the set of target objects, i.e., the target concept (displayed in a green frame for visualization purposes). B: Context-aware - The speaker agent observes the target concept in a context, i.e., a set of target objects (visualized in blue) and a set of distractor objects (visualized in green). This figure was created by the first author.
Fig 3.
Game scenarios added in Experiment 2.
A: Context-unaware + RSA - Speaker agents trained in the context-unaware condition reason about the listener’s likely interpretation of their message. B: Context-aware + RSA - Speaker agents trained in the context-aware condition reason about the listener’s likely interpretation of their message. This figure was created by the first author.
Table 1.
Summary of predictions and measures.
Table 2.
Mean accuracies on the training, validation and test datasets.
Fig 4.
Message lengths per concept hierarchy level.
Concept specificity, or the amount of information that needs to be communicated, increases with the number of fixed attributes that is shared among target objects from more generic concepts (with fewer fixed attributes) on the left, to more specific concepts (with more fixed attributes) on the right.
Table 3.
Mean entropy-based scores, i.e., NMI, effectiveness and consistency.
Fig 5.
NMI, consistency and effectiveness scores for each level of the conceptual hierarchy.
Table 4.
Accuracies on the test dataset.
Fig 6.
Distribution of message lengths for different levels of the conceptual hierarchy.
Table 5.
Lexicon sizes.
Fig 7.
Tradeoff between lexicon size and informativeness.
The lexicon size is normalized by the number of concepts in a dataset.
Fig 8.
Zipf’s law like distribution of message frequency plotted for dataset D(4,4).
D(4,4) is a medium-sized dataset where objects consist of four attributes which can each take four different values.
Fig 9.
Zipf’s law like distribution of message length plotted for dataset D(4,4).
Table 6.
Qualitative examples.
Fig 10.
The three objects in the top row are the target objects that form a target concept together. The three objects in the bottom row are the distractor objects that form the context. A: An example for a specific concept LARGE BLUE CIRCLE in a fine context (two attributes shared). B: An example for a specific concept LARGE BLUE CIRCLE in a coarse context (no attribute shared). C: An example for a generic concept CIRCLE in a coarse context.
Fig 11.
Information-theoretic scores and ambiguity in language.
The set of concepts C is the red circle on the left (red and purple), the set of messages M is the full blue circle on the right (blue and purple). The mutual information I(C,M) (only purple) captures one-to-one mappings between messages and concepts. The conditional entropy (only red) captures many-to-one mappings between messages and concepts. The conditional entropy
(only blue) captures one-to-many mappings between messages and concepts.