LARS-VSA: A Vector Symbolic Architecture For Learning with Abstract Rules

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mejri, Mohamed, Amarnath, Chandramouli, Chatterjee, Abhijit
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916257616363520
author Mejri, Mohamed
Amarnath, Chandramouli
Chatterjee, Abhijit
author_facet Mejri, Mohamed
Amarnath, Chandramouli
Chatterjee, Abhijit
contents Human cognition excels at symbolic reasoning, deducing abstract rules from limited samples. This has been explained using symbolic and connectionist approaches, inspiring the development of a neuro-symbolic architecture that combines both paradigms. In parallel, recent studies have proposed the use of a "relational bottleneck" that separates object-level features from abstract rules, allowing learning from limited amounts of data . While powerful, it is vulnerable to the curse of compositionality meaning that object representations with similar features tend to interfere with each other. In this paper, we leverage hyperdimensional computing, which is inherently robust to such interference to build a compositional architecture. We adapt the "relational bottleneck" strategy to a high-dimensional space, incorporating explicit vector binding operations between symbols and relational representations. Additionally, we design a novel high-dimensional attention mechanism that leverages this relational representation. Our system benefits from the low overhead of operations in hyperdimensional space, making it significantly more efficient than the state of the art when evaluated on a variety of test datasets, while maintaining higher or equal accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2405_14436
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LARS-VSA: A Vector Symbolic Architecture For Learning with Abstract Rules
Mejri, Mohamed
Amarnath, Chandramouli
Chatterjee, Abhijit
Artificial Intelligence
Machine Learning
Human cognition excels at symbolic reasoning, deducing abstract rules from limited samples. This has been explained using symbolic and connectionist approaches, inspiring the development of a neuro-symbolic architecture that combines both paradigms. In parallel, recent studies have proposed the use of a "relational bottleneck" that separates object-level features from abstract rules, allowing learning from limited amounts of data . While powerful, it is vulnerable to the curse of compositionality meaning that object representations with similar features tend to interfere with each other. In this paper, we leverage hyperdimensional computing, which is inherently robust to such interference to build a compositional architecture. We adapt the "relational bottleneck" strategy to a high-dimensional space, incorporating explicit vector binding operations between symbols and relational representations. Additionally, we design a novel high-dimensional attention mechanism that leverages this relational representation. Our system benefits from the low overhead of operations in hyperdimensional space, making it significantly more efficient than the state of the art when evaluated on a variety of test datasets, while maintaining higher or equal accuracy.
title LARS-VSA: A Vector Symbolic Architecture For Learning with Abstract Rules
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2405.14436