English
Related papers

Related papers: Lower bounds for one-layer transformers that compu…

200 papers

Despite their prevalence in neural networks we still lack a thorough theoretical characterization of ReLU layers. This paper aims to further our understanding of ReLU layers by studying how the activation function ReLU interacts with the…

Machine Learning · Computer Science 2019-08-13 Sören Dittmer , Emily J. King , Peter Maass

We provide a systematic recipe for translating ReLU approximation results to softmax attention mechanism. This recipe covers many common approximation targets. Importantly, it yields target-specific, economic resource bounds beyond…

Machine Learning · Computer Science 2026-04-29 Jerry Yao-Chieh Hu , Mingcheng Lu , Yi-Chen Lee , Han Liu

Neural networks activated by the rectified linear unit (ReLU) play a central role in the recent development of deep learning. The topic of approximating functions from H\"older spaces by these networks is crucial for understanding the…

Machine Learning · Computer Science 2023-07-25 Tong Mao , Ding-Xuan Zhou

Neural networks with the Rectified Linear Unit (ReLU) nonlinearity are described by a vector of parameters $\theta$, and realized as a piecewise linear continuous function $R_{\theta}: x \in \mathbb R^{d} \mapsto R_{\theta}(x) \in \mathbb…

Machine Learning · Computer Science 2022-06-08 Pierre Stock , Rémi Gribonval

Quantile regression is the task of estimating a specified percentile response, such as the median, from a collection of known covariates. We study quantile regression with rectified linear unit (ReLU) neural networks as the chosen model…

Statistics Theory · Mathematics 2020-12-21 Oscar Hernan Madrid Padilla , Wesley Tansey , Yanzhen Chen

Attention-based mechanisms are widely used in machine learning, most prominently in transformers. However, hyperparameters such as the rank of the attention matrices and the number of heads are scaled nearly the same way in all realizations…

Machine Learning · Computer Science 2024-07-24 Noah Amsel , Gilad Yehudai , Joan Bruna

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,…

Machine Learning · Computer Science 2023-10-26 Kazuki Irie , Róbert Csordás , Jürgen Schmidhuber

The injectivity of ReLU layers in neural networks, the recovery of vectors from clipped or saturated measurements, and (real) phase retrieval in $\mathbb{R}^n$ allow for a similar problem formulation and characterization using frame theory.…

Machine Learning · Computer Science 2025-02-17 Daniel Freeman , Daniel Haider

Several recent studies have demonstrated that attention-based networks, such as Vision Transformer (ViT), can outperform Convolutional Neural Networks (CNNs) on several computer vision tasks without using convolutional layers. This…

Computer Vision and Pattern Recognition · Computer Science 2021-11-04 Shanda Li , Xiangning Chen , Di He , Cho-Jui Hsieh

Chain-of-thought reasoning and scratchpads have emerged as critical tools for enhancing the computational capabilities of transformers. While theoretical results show that polynomial-length scratchpads can extend transformers' expressivity…

Machine Learning · Computer Science 2025-07-15 Alireza Amiri , Xinting Huang , Mark Rofin , Michael Hahn

Multiplication layers are a key component in various influential neural network modules, including self-attention and hypernetwork layers. In this paper, we investigate the approximation capabilities of deep neural networks with…

Machine Learning · Computer Science 2023-01-12 Ido Ben-Shaul , Tomer Galanti , Shai Dekel

Transformer self-attention can be interpreted as a gradient flow on the unit sphere, in which tokens evolve under softmax interaction potentials and tend to form clusters. While prior work has established clustering behavior for single-head…

Machine Learning · Computer Science 2026-05-11 Ayan Pendharkar

This note provides an interesting observation on casting partial least square (PLS) as a linearized self-attention so that PLS may be studied within the neural network paradigm. On the other hand, the dimensionality reduction and selection…

Machine Learning · Computer Science 2026-05-28 Jiangsheng , You

Large number of ReLU and MAC operations of Deep neural networks make them ill-suited for latency and compute-efficient private inference. In this paper, we present a model optimization method that allows a model to learn to be shallow. In…

Machine Learning · Computer Science 2023-04-27 Souvik Kundu , Yuke Zhang , Dake Chen , Peter A. Beerel

Many empirical studies have demonstrated the performance benefits of conditional computation in neural networks, including reduced inference time and power consumption. We study the fundamental limits of neural conditional computation from…

Machine Learning · Computer Science 2023-03-21 Erdem Koyuncu

Transformer models have achieved remarkable results in a wide range of applications. However, their scalability is hampered by the quadratic time and memory complexity of the self-attention mechanism concerning the sequence length. This…

Machine Learning · Computer Science 2024-02-27 Yury Nahshan , Joseph Kampeas , Emir Haleva

Feature normalization transforms such as Batch and Layer-Normalization have become indispensable ingredients of state-of-the-art deep neural networks. Recent studies on fine-tuning large pretrained models indicate that just tuning the…

Machine Learning · Computer Science 2023-07-06 Angeliki Giannou , Shashank Rajput , Dimitris Papailiopoulos

We derive upper bounds on the complexity of ReLU neural networks approximating the solution maps of parametric partial differential equations. In particular, without any knowledge of its concrete shape, we use the inherent…

Numerical Analysis · Mathematics 2020-05-15 Gitta Kutyniok , Philipp Petersen , Mones Raslan , Reinhold Schneider

We consider functions from the real numbers to the real numbers, output by a neural network with 1 hidden activation layer, arbitrary width, and ReLU activation function. We assume that the parameters of the neural network are chosen…

Machine Learning · Computer Science 2023-04-20 David Holmes

This paper explores the implicit bias of overparameterized neural networks of depth greater than two layers. Our framework considers a family of networks of varying depth that all have the same capacity but different implicitly defined…

Machine Learning · Computer Science 2022-02-03 Greg Ongie , Rebecca Willett
‹ Prev 1 3 4 5 6 7 10 Next ›