English
Related papers

Related papers: The Disharmony between BN and ReLU Causes Gradient…

200 papers

Batch Normalization (BN) is a commonly used technique to accelerate and stabilize training of deep neural networks. Despite its empirical success, a full theoretical understanding of BN is yet to be developed. In this work, we analyze BN…

Machine Learning · Computer Science 2022-03-22 Tolga Ergen , Arda Sahiner , Batu Ozturkler , John Pauly , Morteza Mardani , Mert Pilanci

Researchers have proposed various activation functions. These activation functions help the deep network to learn non-linear behavior with a significant effect on training dynamics and task performance. The performance of these activations…

Computer Vision and Pattern Recognition · Computer Science 2020-01-07 Pravendra Singh , Munender Varshney , Vinay P. Namboodiri

Binarized Neural Networks (BNNs) can significantly reduce the inference latency and energy consumption in resource-constrained devices due to their pure-logical computation and fewer memory accesses. However, training BNNs is difficult…

Computer Vision and Pattern Recognition · Computer Science 2019-04-08 Ruizhou Ding , Ting-Wu Chin , Zeye Liu , Diana Marculescu

Intriguing empirical evidence exists that deep learning can work well with exoticschedules for varying the learning rate. This paper suggests that the phenomenon may be due to Batch Normalization or BN, which is ubiquitous and provides…

Machine Learning · Computer Science 2019-11-22 Zhiyuan Li , Sanjeev Arora

Deep neural networks (DNNs) are playing key roles in various artificial intelligence applications such as image classification and object recognition. However, a growing number of studies have shown that there exist adversarial examples in…

Machine Learning · Computer Science 2020-10-22 Hongying Liu , Zhenyu Zhou , Fanhua Shang , Xiaoyu Qi , Yuanyuan Liu , Licheng Jiao

In this work, we investigate a particular implicit bias in gradient descent training, which we term "Feature Averaging," and argue that it is one of the principal factors contributing to the non-robustness of deep neural networks. We show…

Machine Learning · Computer Science 2025-03-04 Binghui Li , Zhixuan Pan , Kaifeng Lyu , Jian Li

There are two widely known issues with properly training Recurrent Neural Networks, the vanishing and the exploding gradient problems detailed in Bengio et al. (1994). In this paper we attempt to improve the understanding of the underlying…

Machine Learning · Computer Science 2013-02-19 Razvan Pascanu , Tomas Mikolov , Yoshua Bengio

Vanishing (and exploding) gradients effect is a common problem for recurrent neural networks with nonlinear activation functions which use backpropagation method for calculation of derivatives. Deep feedforward neural networks with many…

Neural and Evolutionary Computing · Computer Science 2017-02-15 Artem Chernodub , Dimitri Nowicki

Solving mixed-integer optimization problems with embedded neural networks with ReLU activation functions is challenging. Big-M coefficients that arise in relaxing binary decisions related to these functions grow exponentially with the…

Optimization and Control · Mathematics 2025-02-06 Christoph Plate , Mirko Hahn , Alexander Klimek , Caroline Ganzer , Kai Sundmacher , Sebastian Sager

The choice of activation function plays a critical role in neural networks, yet most architectures still rely on fixed, uniform activation functions across all neurons. We introduce SmartMixed, a two-phase training strategy that allows…

Machine Learning · Computer Science 2025-11-03 Amin Omidvar

Reinforcement learning (RL) is a dominant paradigm for training autonomous agents, yet these agents often exhibit poor generalization, failing to adapt to scenarios not seen during training. In this work, we identify a fundamental cause of…

Artificial Intelligence · Computer Science 2026-01-16 Jingyu Liu , Xiaopeng Wu , Jingquan Peng , Kehan Chen , Chuan Yu , Lizhong Ding , Yong Liu

Generative Adversarial Imitation Learning (GAIL) stands as a cornerstone approach in imitation learning. This paper investigates the gradient explosion in two types of GAIL: GAIL with deterministic policy (DE-GAIL) and GAIL with stochastic…

Machine Learning · Computer Science 2023-12-19 Wanying Wang , Yichen Zhu , Yirui Zhou , Chaomin Shen , Jian Tang , Zhiyuan Xu , Yaxin Peng , Yangchun Zhang

We identify and study two common failure modes for early training in deep ReLU nets. For each we give a rigorous proof of when it occurs and how to avoid it, for fully connected and residual architectures. The first failure mode,…

Machine Learning · Statistics 2018-11-14 Boris Hanin , David Rolnick

Machine Unlearning has recently garnered significant attention, aiming to selectively remove knowledge associated with specific data while preserving the model's performance on the remaining data. A fundamental challenge in this process is…

Machine Learning · Computer Science 2025-07-29 Gaurav Patel , Qiang Qiu

We introduce a variational framework to learn the activation functions of deep neural networks. Our aim is to increase the capacity of the network while controlling an upper-bound of the actual Lipschitz constant of the input-output…

Machine Learning · Computer Science 2023-07-19 Shayan Aziznejad , Harshit Gupta , Joaquim Campos , Michael Unser

The impact of gradient noise on training deep models is widely acknowledged but not well understood. In this context, we study the distribution of gradients during training. We introduce a method, Gradient Clustering, to minimize the…

Machine Learning · Computer Science 2020-07-10 Fartash Faghri , David Duvenaud , David J. Fleet , Jimmy Ba

Deep learning empirically achieves high performance in many applications, but its training dynamics has not been fully understood theoretically. In this paper, we explore theoretical analysis on training two-layer ReLU neural networks in a…

Machine Learning · Statistics 2021-06-30 Shunta Akiyama , Taiji Suzuki

Nonlinear activation functions are widely recognized for enhancing the expressivity of neural networks, which is the primary reason for their widespread implementation. In this work, we focus on ReLU activation and reveal a novel and…

Machine Learning · Computer Science 2025-10-22 Chaoyue Liu , Han Bi , Like Hui , Xiao Liu

Training instability in modern deep learning systems is frequently triggered by rare but extreme gradient-norm spikes, which can induce oversized parameter updates, corrupt optimizer state, and lead to slow recovery or divergence. Widely…

Amongst others, the adoption of Rectified Linear Units (ReLUs) is regarded as one of the ingredients of the success of deep learning. ReLU activation has been shown to mitigate the vanishing gradient issue, to encourage sparsity in the…

Machine Learning · Statistics 2021-10-14 Nicola Picchiotti , Marco Gori