English
Related papers

Related papers: Sparse Growing Transformer: Training-Time Sparse D…

200 papers

Structured pruning of Generative Pre-trained Transformers (GPTs) offers a promising path to efficiency but often suffers from irreversible performance degradation due to the discarding of transformer blocks. In this paper, we introduce…

Machine Learning · Computer Science 2025-12-16 Zehua Pei , Hui-Ling Zhen , Xianzhi Yu , Sinno Jialin Pan , Mingxuan Yuan , Bei Yu

Training sparse networks to converge to the same performance as dense neural architectures has proven to be elusive. Recent work suggests that initialization is the key. However, while this direction of research has had some success,…

Machine Learning · Computer Science 2021-06-17 Kale-ab Tessera , Sara Hooker , Benjamin Rosman

Modern deep learning (DL) architectures are trained using variants of the SGD algorithm that is run with a $\textit{manually}$ defined learning rate schedule, i.e., the learning rate is dropped at the pre-defined epochs, typically when the…

Machine Learning · Computer Science 2021-12-14 Yunfei Teng , Jing Wang , Anna Choromanska

Diffusion Transformers (DiTs) dominate video generation but their high computational cost severely limits real-world applicability, usually requiring tens of minutes to generate a few seconds of video even on high-performance GPUs. This…

Computer Vision and Pattern Recognition · Computer Science 2025-04-29 Haocheng Xi , Shuo Yang , Yilong Zhao , Chenfeng Xu , Muyang Li , Xiuyu Li , Yujun Lin , Han Cai , Jintao Zhang , Dacheng Li , Jianfei Chen , Ion Stoica , Kurt Keutzer , Song Han

In this work, we propose a Bayesian type sparse deep learning algorithm. The algorithm utilizes a set of spike-and-slab priors for the parameters in the deep neural network. The hierarchical Bayesian mixture will be trained using an…

Numerical Analysis · Mathematics 2021-03-17 Yating Wang , Wei Deng , Lin Guang

The growing energy and performance costs of deep learning have driven the community to reduce the size of neural networks by selectively pruning components. Similarly to their biological counterparts, sparse networks generalize just as…

Machine Learning · Computer Science 2021-02-02 Torsten Hoefler , Dan Alistarh , Tal Ben-Nun , Nikoli Dryden , Alexandra Peste

Sparse training is emerging as a promising avenue for reducing the computational cost of training neural networks. Several recent studies have proposed pruning methods using learnable thresholds to efficiently explore the non-uniform…

Machine Learning · Computer Science 2023-04-17 Abhisek Kundu , Naveen K. Mellempudi , Dharma Teja Vooturi , Bharat Kaul , Pradeep Dubey

Pre-trained Transformers inherently possess the characteristic of sparse activation, where only a small fraction of the neurons are activated for each token. While sparse activation has been explored through post-training methods, its…

Computation and Language · Computer Science 2024-10-07 Zhengyan Zhang , Chaojun Xiao , Qiujieli Qin , Yankai Lin , Zhiyuan Zeng , Xu Han , Zhiyuan Liu , Ruobing Xie , Maosong Sun , Jie Zhou

Deep reinforcement learning (DRL) has shown remarkable success in sequential decision-making problems but suffers from a long training time to obtain such good performance. Many parallel and distributed DRL training approaches have been…

Machine Learning · Computer Science 2021-01-26 Juhyoung Lee , Sangyeob Kim , Sangjin Kim , Wooyoung Jo , Hoi-Jun Yoo

The success of self-attention lies in its ability to capture long-range dependencies and enhance context understanding, but it is limited by its computational complexity and challenges in handling sequential data with inherent…

Computation and Language · Computer Science 2025-05-05 Md Kowsher , Nusrat Jahan Prottasha , Chun-Nam Yu , Ozlem Ozmen Garibay , Niloofar Yousefi

Structured sparsity accelerates training and inference on modern GPUs, yet it still trails unstructured dynamic sparse training (DST) in accuracy. The shortfall stems from a loss of expressivity: whereas a dense layer can realize every…

Machine Learning · Computer Science 2025-10-17 Abhishek Tyagi , Arjun Iyer , Liam Young , William H Renninger , Christopher Kanan , Yuhao Zhu

To train deep learning models faster, distributed training on multiple GPUs is the very popular scheme in recent years. However, the communication bandwidth is still a major bottleneck of training performance. To improve overall training…

Machine Learning · Computer Science 2022-09-20 Daegun Yoon , Sangyoon Oh

Deep Gaussian Processes (DGPs) are multi-layer, flexible extensions of Gaussian processes but their training remains challenging. Sparse approximations simplify the training but often require optimization over a large number of inducing…

Machine Learning · Statistics 2021-07-20 Ayush Jain , P. K. Srijith , Mohammad Emtiyaz Khan

Deep neural networks (DNNs) are effective in solving many real-world problems. Larger DNN models usually exhibit better quality (e.g., accuracy) but their excessive computation results in long inference time. Model sparsification can reduce…

Computer Vision and Pattern Recognition · Computer Science 2022-03-07 Xiaolong Ma , Minghai Qin , Fei Sun , Zejiang Hou , Kun Yuan , Yi Xu , Yanzhi Wang , Yen-Kuang Chen , Rong Jin , Yuan Xie

Spiking Neural Networks (SNNs) are more biologically plausible and computationally efficient. Therefore, SNNs have the natural advantage of drawing the sparse structural plasticity of brain development to alleviate the energy problems of…

Neural and Evolutionary Computing · Computer Science 2023-02-06 Bing Han , Feifei Zhao , Yi Zeng , Wenxuan Pan

The remarkable capabilities of modern large reasoning models are largely unlocked through post-training techniques such as supervised fine-tuning (SFT) and reinforcement learning (RL). However, the architectural mechanisms behind such…

Artificial Intelligence · Computer Science 2026-04-15 Yein Park , Minbyul Jeong , Jaewoo Kang

A major drawback of backpropagation through time (BPTT) is the difficulty of learning long-term dependencies, coming from having to propagate credit information backwards through every single step of the forward computation. This makes BPTT…

Artificial Intelligence · Computer Science 2017-11-08 Nan Rosemary Ke , Anirudh Goyal , Olexa Bilaniuk , Jonathan Binas , Laurent Charlin , Chris Pal , Yoshua Bengio

Sparse attention reduces the quadratic complexity of full self-attention but faces two challenges: (1) an attention gap, where applying sparse attention to full-attention-trained models causes performance degradation due to train-inference…

Computation and Language · Computer Science 2026-02-02 Zhenyi Shen , Junru Lu , Lin Gui , Jiazheng Li , Yulan He , Di Yin , Xing Sun

Deep neural networks (DNNs) have been proven to be effective in solving many real-life problems, but its high computation cost prohibits those models from being deployed to edge devices. Pruning, as a method to introduce zeros to model…

Machine Learning · Computer Science 2021-12-22 Fei Sun , Minghai Qin , Tianyun Zhang , Xiaolong Ma , Haoran Li , Junwen Luo , Zihao Zhao , Yen-Kuang Chen , Yuan Xie

The structural re-parameterization (SRP) technique is a novel deep learning technique that achieves interconversion between different network architectures through equivalent parameter transformations. This technique enables the mitigation…

Computer Vision and Pattern Recognition · Computer Science 2024-08-08 Shanshan Zhong , Zhongzhan Huang , Wushao Wen , Jinghui Qin , Liang Lin