paper-with-me

홈 › Papers

Token Space: A Category Theory Framework for AI Computations

2024-04-11 · Wuming Pan

This paper introduces the Token Space framework, a novel mathematical construct designed to enhance the interpretability and effectiveness of deep learning models through the application of category theory. By establishing a categorical structure at the Token level, we provide a new lens through which AI computations can be understood, emphasizing the relationships between tokens, such as grouping, order, and parameter types. We explore the foundational methodologies of the Token Space, detailing its construction, the role of construction operators and initial categories, and its application in analyzing deep learning models, specifically focusing on attention mechanisms and Transformer architectures. The integration of category theory into AI research offers a unified framework to describe and analyze computational structures, enabling new research paths and development possibilities. Our investigation reveals that the Token Space framework not only facilitates a deeper theoretical understanding of deep learning models but also opens avenues for the design of more efficient, interpretable, and innovative models, illustrating the significant role of category theory in advancing computational models.

📄 PDF Abstract BibTeX arXiv:2404.11624

Code (0)

등록된 구현이 없습니다.

Tasks

Deep Learning

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Adam 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Multi-Head Attention 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Position-Wise Feed-Forward Layer 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…

Similar Papers 제목 키워드 기반

Streaming Attention Approximation via Discrepancy Theory

2025-02-11 · Insu Han, Michael Kapralov, Ekaterina Kochetkova, Kshiteej Sheth 외

Large language models (LLMs) have achieved impressive success, but their high memory requirements present challenges for long-context token generation. In this paper we study the streaming complexity of attention approxi…

Differentiable Causal Computations via Delayed Trace

2019-03-04 · David Sprunger, Shin-ya Katsumata

We investigate causal computations taking sequences of inputs to sequences of outputs where the $n$th output depends on the first $n$ inputs only. We model these in category theory via a construction taking a Cartesian c…

Rolling Shutter Correction

Topos Theory for Generative AI and LLMs

2025-08-05 · Sridhar Mahadevan arxiv

We propose the design of novel categorical generative AI architectures (GAIAs) using topos theory, a type of category that is ``set-like": a topos has all (co)limits, is Cartesian closed, and has a subobject classifier. …

A Convenient Category for Higher-Order Probability Theory

2017-01-10 · Chris Heunen, Ohad Kammar, Sam Staton, Hongseok Yang

Higher-order probabilistic programming languages allow programmers to write sophisticated models in machine learning and statistics in a succinct and structured way, but step outside the standard measure-theoretic formal…

FormProbabilistic Programming

Neural Collapse Is Forbidden: Information Floors in Language Models

2026-07-10 · Bruno Abrahao arxiv

Within-class variance in language-model representations is commonly read as incomplete neural collapse. We argue it is allocated information storage, and that the allocation obeys a law. A one-line centering identity voi…