Learning a Fourier Transform for Linear Relative Positional Encodings in Transformers

• URL: http://arxiv.org/abs/2302.01925v2
• Date: Wed, 3 Apr 2024 21:24:50 GMT
• Title: Learning a Fourier Transform for Linear Relative Positional Encodings in Transformers
• Authors: Krzysztof Marcin Choromanski, Shanda Li, Valerii Likhosherstov, Kumar Avinava Dubey, Shengjie Luo, Di He, Yiming Yang, Tamas Sarlos, Thomas Weingarten, Adrian Weller,
• Abstract summary: We propose a new class of linear Transformers calledLearner-Transformers (Learners) They incorporate a wide range of relative positional encoding mechanisms (RPEs) These include regular RPE techniques applied for sequential data, as well as novel RPEs operating on geometric data embedded in higher-dimensional Euclidean spaces.
• Score: 71.32827362323205
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We propose a new class of linear Transformers called FourierLearner-Transformers (FLTs), which incorporate a wide range of relative positional encoding mechanisms (RPEs). These include regular RPE techniques applied for sequential data, as well as novel RPEs operating on geometric data embedded in higher-dimensional Euclidean spaces. FLTs construct the optimal RPE mechanism implicitly by learning its spectral representation. As opposed to other architectures combining efficient low-rank linear attention with RPEs, FLTs remain practical in terms of their memory usage and do not require additional assumptions about the structure of the RPE mask. Besides, FLTs allow for applying certain structural inductive bias techniques to specify masking strategies, e.g. they provide a way to learn the so-called local RPEs introduced in this paper and give accuracy gains as compared with several other linear Transformers for language modeling. We also thoroughly test FLTs on other data modalities and tasks, such as image classification, 3D molecular modeling, and learnable optimizers. To the best of our knowledge, for 3D molecular data, FLTs are the first Transformer architectures providing linear attention and incorporating RPE masking.

Related papers

• Re-Parameterization of Lightweight Transformer for On-Device Speech Emotion Recognition [10.302458835329539]
  We introduce a new method, namely Transformer Re- parameterization, to boost the performance of lightweight Transformer models. Experimental results show that our proposed method consistently improves the performance of lightweight Transformers, even making them comparable to large models.
  arXiv  Detail & Related papers  (2024-11-14T10:36:19Z)
• Learning Linear Attention in Polynomial Time [115.68795790532289]
  We provide the first results on learnability of single-layer Transformers with linear attention. We show that linear attention may be viewed as a linear predictor in a suitably defined RKHS. We show how to efficiently identify training datasets for which every empirical riskr is equivalent to the linear Transformer.
  arXiv  Detail & Related papers  (2024-10-14T02:41:01Z)
• Flatten Anything: Unsupervised Neural Surface Parameterization [76.4422287292541]
  We introduce the Flatten Anything Model (FAM), an unsupervised neural architecture to achieve global free-boundary surface parameterization. Compared with previous methods, our FAM directly operates on discrete surface points without utilizing connectivity information. Our FAM is fully-automated without the need for pre-cutting and can deal with highly-complex topologies.
  arXiv  Detail & Related papers  (2024-05-23T14:39:52Z)
• Self-Supervised Pre-Training for Table Structure Recognition Transformer [25.04573593082671]
  We propose a self-supervised pre-training (SSP) method for table structure recognition transformers. We discover that the performance gap between the linear projection transformer and the hybrid CNN-transformer can be mitigated by SSP of the visual encoder in the TSR model.
  arXiv  Detail & Related papers  (2024-02-23T19:34:06Z)
• How Do Transformers Learn In-Context Beyond Simple Functions? A Case Study on Learning with Representations [98.7450564309923]
  This paper takes initial steps on understanding in-context learning (ICL) in more complex scenarios, by studying learning with representations. We construct synthetic in-context learning problems with a compositional structure, where the label depends on the input through a possibly complex but fixed representation function. We show theoretically the existence of transformers that approximately implement such algorithms with mild depth and size.
  arXiv  Detail & Related papers  (2023-10-16T17:40:49Z)
• Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selection [88.23337313766353]
  This work first provides a comprehensive statistical theory for transformers to perform ICL. We show that transformers can implement a broad class of standard machine learning algorithms in context. A emphsingle transformer can adaptively select different base ICL algorithms.
  arXiv  Detail & Related papers  (2023-06-07T17:59:31Z)
• Your Transformer May Not be as Powerful as You Expect [88.11364619182773]
  We mathematically analyze the power of RPE-based Transformers regarding whether the model is capable of approximating any continuous sequence-to-sequence functions. We present a negative result by showing there exist continuous sequence-to-sequence functions that RPE-based Transformers cannot approximate no matter how deep and wide the neural network is. We develop a novel attention module, called Universal RPE-based (URPE) Attention, which satisfies the conditions.
  arXiv  Detail & Related papers  (2022-05-26T14:51:30Z)
• Relative Positional Encoding for Transformers with Linear Complexity [30.48367640796256]
  relative positional encoding (RPE) was proposed as beneficial for classical Transformers. RPE is not available for the recent linear-variants of the Transformer, because it requires the explicit computation of the attention matrix. In this paper, we present precisely what is precisely what is a way to generate PE that can be used as a replacement to the classical additive (sinusoidal) PE and provably behaves like RPE.
  arXiv  Detail & Related papers  (2021-05-18T09:52:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.