REVIEW 3 cited by
Transformers are Deep Infinite-Dimensional Non-Mercer Binary Kernel Machines
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Despite their ubiquity in core AI fields like natural language processing, the mechanics of deep attention-based neural networks like the Transformer model are not fully understood. In this article, we present a new perspective towards understanding how Transformers work. In particular, we show that the "dot-product attention" that is the core of the Transformer's operation can be characterized as a kernel learning method on a pair of Banach spaces. In particular, the Transformer's kernel is characterized as having an infinite feature dimension. Along the way we consider an extension of the standard kernel learning problem to a binary setting, where data come from two input domains and a response is defined for every cross-domain pair. We prove a new representer theorem for these binary kernel machines with non-Mercer (indefinite, asymmetric) kernels (implying that the functions learned are elements of reproducing kernel Banach spaces rather than Hilbert spaces), and also prove a new universal approximation theorem showing that the Transformer calculation can learn any binary non-Mercer reproducing kernel Banach space pair. We experiment with new kernels in Transformers, and obtain results that suggest the infinite dimensionality of the standard Transformer kernel is partially responsible for its performance. This paper's results provide a new theoretical understanding of a very important but poorly understood model in modern machine~learning.
Forward citations
Cited by 3 Pith papers
-
From Self-Attention to Connection Laplacian: A Unified Operator View of Transformers
Multi-head attention is exactly a scaled edge-dependent connection walk, and trained Transformers empirically develop stable walks and approximate scaled-isometric transports that strengthen with scale.
-
Tokenizing Electron Cloud in Protein-Ligand Interaction Learning
ECBind tokenizes electron cloud densities via quantized embeddings and improves protein-ligand binding affinity prediction, especially per-structure correlations on MISATO.
-
On the Complexity-Faithfulness Trade-off of Gradient-Based Explanations
The paper introduces EF and ΔEF as spectral metrics, but ΔEF is derived from EF, making the complexity-faithfulness trade-off partly tautological.
Discussion (0). Sign in to comment.