Emergence of Separable Manifolds in Deep Language Representations

Abstract

Deep neural networks (DNNs) have shown much empirical success in solving perceptual tasks across various cognitive modalities. While they are only loosely inspired by the biological brain, recent studies report considerable similarities between representations extracted from task-optimized DNNs and neural populations in the brain. DNNs have subsequently become a popular model class to infer computational principles underlying complex cognitive functions, and in turn, they have also emerged as a natural testbed for applying methods originally developed to probe information in neural populations. In this work, we utilize mean-field theoretic manifold analysis, a recent technique from computational neuroscience that connects geometry of feature representations with linear separability of classes, to analyze language representations from large-scale contextual embedding models. We explore representations from different model families (BERT, RoBERTa, GPT, etc.) and find evidence for emergence of linguistic manifolds across layer depth (e.g., manifolds for part-of-speech tags), especially in ambiguous data (i.e, words with multiple part-of-speech tags, or part-of-speech classes including many words). In addition, we find that the emergence of linear separability in these manifolds is driven by a combined reduction of manifolds' radius, dimensionality and inter-manifold correlations.

arXiv comment9 pages. 10 figures. Accepted to ICML 2020. Included supplemental materials

References in corpus (7)

• SVCCA: Singular Vector Canonical Correlation Analysis for Deep Learning Dynamics and Interpretability
• Classification and Geometry of General Perceptual Manifolds
• Semantic Tagging with Deep Residual Networks
• Language Modeling Teaches You More Syntax than Translation Does: Lessons Learned Through Auxiliary Task Analysis
• Dimensionality compression and expansion in Deep Neural Networks
• Geodesics of learned representations
• Untangling in Invariant Speech Recognition

Cited by in corpus (3)

• On the geometry of generalization and memorization in deep neural networks
• On convex decision regions in deep network representations
• Design considerations for a hierarchical semantic compositional framework for medical natural language understanding