paper-with-me

홈 › Papers

Practical Computational Power of Linear Transformers and Their Recurrent and Self-Referential Extensions

2023-10-24 · Kazuki Irie, Róbert Csordás, Jürgen Schmidhuber

Recent studies of the computational power of recurrent neural networks (RNNs) reveal a hierarchy of RNN architectures, given real-time and finite-precision assumptions. Here we study auto-regressive Transformers with linearised attention, a.k.a. linear Transformers (LTs) or Fast Weight Programmers (FWPs). LTs are special in the sense that they are equivalent to RNN-like sequence processors with a fixed-size state, while they can also be expressed as the now-popular self-attention networks. We show that many well-known results for the standard Transformer directly transfer to LTs/FWPs. Our formal language recognition experiments demonstrate how recently proposed FWP extensions such as recurrent FWPs and self-referential weight matrices successfully overcome certain limitations of the LT, e.g., allowing for generalisation on the parity problem. Our code is public.

📄 PDF Abstract BibTeX arXiv:2310.16076

Code (1)

idsia/fwp-formal-lang 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Adam 설명 없음
Position-Wise Feed-Forward Layer 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Residual Connection 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

k-Maximum Inner Product Attention for Graph Transformers and the Expressive Power of GraphGPS

2026-04-04 · Jonas De Schouwer, Haitz Sáez de Ocáriz Borde, Xiaowen Dong arxiv

Graph transformers have shown promise in overcoming limitations of traditional graph neural networks, such as oversquashing and difficulties in modeling long-range dependencies. However, their application to large-scale …

On the Computational Power of Transformers and its Implications in Sequence Modeling

2020-06-16 · CONLL 2020 · Satwik Bhattamishra, Arkil Patel, Navin Goyal

Transformers are being used extensively across several sequence modeling tasks. Significant research effort has been devoted to experimentally probe the inner workings of Transformers. However, our conceptual and theoret…

Machine TranslationTranslation

PEANO-ViT: Power-Efficient Approximations of Non-Linearities in Vision Transformers

2024-06-21 · Mohammad Erfan Sadeghi, Arash Fayyazi, Seyedarmin Azizi, Massoud Pedram

The deployment of Vision Transformers (ViTs) on hardware platforms, specially Field-Programmable Gate Arrays (FPGAs), presents many challenges, which are mainly due to the substantial computational and power requirements…

Conformal Transformations for Symmetric Power Transformers

2025-03-05 · Saurabh Kumar, Jacob Buckman, Carles Gelada, Sean Zhang

Transformers with linear attention offer significant computational advantages over softmax-based transformers but often suffer from degraded performance. The symmetric power (sympow) transformer, a particular type of lin…

The Parallelism Tradeoff: Limitations of Log-Precision Transformers

2022-07-02 · William Merrill, Ashish Sabharwal

Despite their omnipresence in modern NLP, characterizing the computational power of transformer neural nets remains an interesting open question. We prove that transformers whose arithmetic precision is logarithmic in th…

Open-Ended Question Answering