Breaking the Conventional Forward-Backward Tie in Neural Networks: Activation Functions

• URL: http://arxiv.org/abs/2509.07236v1
• Date: Mon, 08 Sep 2025 21:30:00 GMT
• Title: Breaking the Conventional Forward-Backward Tie in Neural Networks: Activation Functions
• Authors: Luigi Troiano, Francesco Gissi, Vincenzo Benedetto, Genny Tortora,
• Abstract summary: We show that precise gradient magnitudes derived from activation functions are largely redundant, provided the gradient direction is preserved.<n>We explicitly demonstrate that neural networks with non-differentiable activation functions, such as the Heaviside step function, can be effectively trained.
• Score: 0.1633272850273525
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Gradient-based neural network training traditionally enforces symmetry between forward and backward propagation, requiring activation functions to be differentiable (or sub-differentiable) and strictly monotonic in certain regions to prevent flat gradient areas. This symmetry, linking forward activations closely to backward gradients, significantly restricts the selection of activation functions, particularly excluding those with substantial flat or non-differentiable regions. In this paper, we challenge this assumption through mathematical analysis, demonstrating that precise gradient magnitudes derived from activation functions are largely redundant, provided the gradient direction is preserved. Empirical experiments conducted on foundational architectures - such as Multi-Layer Perceptrons (MLPs), Convolutional Neural Networks (CNNs), and Binary Neural Networks (BNNs) - confirm that relaxing forward-backward symmetry and substituting traditional gradients with simpler or stochastic alternatives does not impair learning and may even enhance training stability and efficiency. We explicitly demonstrate that neural networks with flat or non-differentiable activation functions, such as the Heaviside step function, can be effectively trained, thereby expanding design flexibility and computational efficiency. Further empirical validation with more complex architectures remains a valuable direction for future research.

Related papers

• The Butterfly Effect: Neural Network Training Trajectories Are Highly Sensitive to Initial Conditions [51.68215326304272]
  We show that even small perturbations reliably cause otherwise identical training trajectories to diverge-an effect that diminishes rapidly over training time.<n>Our findings provide insights into neural network training stability, with practical implications for fine-tuning, model merging, and diversity of model ensembles.
  arXiv  Detail & Related papers  (2025-06-16T08:35:16Z)
• ResQuNNs: Towards Enabling Deep Learning in Quantum Convolution Neural Networks [4.348591076994875]
  We present a novel framework for enhancing the performance of Quanvolutional Neural Networks (QuNNs) by introducing trainable quanvolutional layers.<n>Our research overcomes this limitation by enabling training within these layers, significantly increasing the flexibility and potential of QuNNs.<n>We propose a novel architecture, Residual Quanvolutional Neural Networks (ResQuNNs), leveraging the concept of residual learning.
  arXiv  Detail & Related papers  (2024-02-14T12:55:28Z)
• Feature Mapping in Physics-Informed Neural Networks (PINNs) [1.9819034119774483]
  We study the training dynamics of PINNs with a feature mapping layer via the limiting Conjugate Kernel and Neural Tangent Kernel. We propose conditionally positive definite Radial Basis Function as a better alternative.
  arXiv  Detail & Related papers  (2024-02-10T13:51:09Z)
• Efficient and Flexible Neural Network Training through Layer-wise Feedback Propagation [49.44309457870649]
  Layer-wise Feedback feedback (LFP) is a novel training principle for neural network-like predictors.<n>LFP decomposes a reward to individual neurons based on their respective contributions.<n>Our method then implements a greedy reinforcing approach helpful parts of the network and weakening harmful ones.
  arXiv  Detail & Related papers  (2023-08-23T10:48:28Z)
• ENN: A Neural Network with DCT Adaptive Activation Functions [2.2713084727838115]
  We present Expressive Neural Network (ENN), a novel model in which the non-linear activation functions are modeled using the Discrete Cosine Transform (DCT) This parametrization keeps the number of trainable parameters low, is appropriate for gradient-based schemes, and adapts to different learning tasks. The performance of ENN outperforms state of the art benchmarks, providing above a 40% gap in accuracy in some scenarios.
  arXiv  Detail & Related papers  (2023-07-02T21:46:30Z)
• TANGOS: Regularizing Tabular Neural Networks through Gradient Orthogonalization and Specialization [69.80141512683254]
  We introduce Tabular Neural Gradient Orthogonalization and gradient (TANGOS) TANGOS is a novel framework for regularization in the tabular setting built on latent unit attributions. We demonstrate that our approach can lead to improved out-of-sample generalization performance, outperforming other popular regularization methods.
  arXiv  Detail & Related papers  (2023-03-09T18:57:13Z)
• Globally Optimal Training of Neural Networks with Threshold Activation Functions [63.03759813952481]
  We study weight decay regularized training problems of deep neural networks with threshold activations. We derive a simplified convex optimization formulation when the dataset can be shattered at a certain layer of the network.
  arXiv  Detail & Related papers  (2023-03-06T18:59:13Z)
• Implicit Stochastic Gradient Descent for Training Physics-informed Neural Networks [51.92362217307946]
  Physics-informed neural networks (PINNs) have effectively been demonstrated in solving forward and inverse differential equation problems. PINNs are trapped in training failures when the target functions to be approximated exhibit high-frequency or multi-scale features. In this paper, we propose to employ implicit gradient descent (ISGD) method to train PINNs for improving the stability of training process.
  arXiv  Detail & Related papers  (2023-03-03T08:17:47Z)
• Exploring Linear Feature Disentanglement For Neural Networks [63.20827189693117]
  Non-linear activation functions, e.g., Sigmoid, ReLU, and Tanh, have achieved great success in neural networks (NNs) Due to the complex non-linear characteristic of samples, the objective of those activation functions is to project samples from their original feature space to a linear separable feature space. This phenomenon ignites our interest in exploring whether all features need to be transformed by all non-linear functions in current typical NNs.
  arXiv  Detail & Related papers  (2022-03-22T13:09:17Z)
• Analytical aspects of non-differentiable neural networks [0.0]
  We discuss the expressivity of quantized neural networks and approximation techniques for non-differentiable networks. We show that QNNs have the same expressivity as DNNs in terms of approximation of Lipschitz functions in the $Linfty$ norm. We also consider networks defined by means of Heaviside-type activation functions, and prove for them a pointwise approximation result by means of smooth networks.
  arXiv  Detail & Related papers  (2020-11-03T17:20:43Z)
• Investigating the interaction between gradient-only line searches and different activation functions [0.0]
  Gradient-only line searches (GOLS) adaptively determine step sizes along search directions for discontinuous loss functions in neural network training. We find that GOLS are robust for a range of activation functions, but sensitive to the Rectified Linear Unit (ReLU) activation function in standard feedforward architectures.
  arXiv  Detail & Related papers  (2020-02-23T12:28:27Z)

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.