Analyzing the Capacity of Distributed Vector Representations to Encode Spatial Information
This work addresses a fundamental limitation in cognitive modeling for researchers in AI and cognitive science, though it appears incremental as it builds on existing Vector Symbolic Architectures.
The paper analyzed the capacity limits of distributed vector representations to encode spatial information, finding upper bounds for the number of concepts that can be stored in a single vector.
Vector Symbolic Architectures belong to a family of related cognitive modeling approaches that encode symbols and structures in high-dimensional vectors. Similar to human subjects, whose capacity to process and store information or concepts in short-term memory is subject to numerical restrictions,the capacity of information that can be encoded in such vector representations is limited and one way of modeling the numerical restrictions to cognition. In this paper, we analyze these limits regarding information capacity of distributed representations. We focus our analysis on simple superposition and more complex, structured representations involving convolutive powers to encode spatial information. In two experiments, we find upper bounds for the number of concepts that can effectively be stored in a single vector.