Geometric Monomial (GEM): a family of rational 2N-differentiable activation functions

• URL: http://arxiv.org/abs/2604.21677v1
• Date: Thu, 23 Apr 2026 13:42:49 GMT
• Title: Geometric Monomial (GEM): a family of rational 2N-differentiable activation functions
• Abstract summary: We propose a family of $C2N$-smooth activation functions, achieving ReLU-like performance with purely rational arithmetic.<n>We introduce three variants: GEM (the base family), E-GEM (an $$- parameterized generalization enabling arbitrary $Lp$-approximation of ReLU), and SE-GEM (a piecewise variant eliminating dead neurons with $C2N$ junction smoothness)
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The choice of activation function plays a crucial role in the optimization and performance of deep neural networks. While the Rectified Linear Unit (ReLU) remains the dominant choice due to its simplicity and effectiveness, its lack of smoothness may hinder gradient-based optimization in deep architectures. In this work we propose a family of $C^{2N}$-smooth activation functions whose gate follows a log-logistic CDF, achieving ReLU-like performance with purely rational arithmetic. We introduce three variants: GEM (the base family), E-GEM (an $ε$-parameterized generalization enabling arbitrary $L^p$-approximation of ReLU), and SE-GEM (a piecewise variant eliminating dead neurons with $C^{2N}$ junction smoothness). An $N$-ablation study establishes $N=1$ as optimal for standard-depth networks, reducing the GELU deficit on CIFAR-100 + ResNet-56 from 6.10% to 2.12%. The smoothness parameter $N$ further reveals a CNN-transformer tradeoff: $N=1$ is preferred for deep CNNs, while $N=2$ is preferred for transformers. On MNIST, E-GEM ties the best baseline (99.23%). On CIFAR-10 + ResNet-56, SE-GEM ($ε=10^{-4}$) surpasses GELU (92.51% vs 92.44%) -- the first GEM-family activation to outperform GELU. On CIFAR-100 + ResNet-56, E-GEM reduces the GELU deficit from 6.10% (GEM $N=2$) to just 0.62%. On GPT-2 (124M), GEM achieves the lowest perplexity (72.57 vs 73.76 for GELU), with GEM $N=1$ also beating GELU (73.32). On BERT-small, E-GEM ($ε=10$) achieves the best validation loss (6.656) across all activations. The $ε$-parameterization reveals a scale-dependent optimum: small $ε$ ($10^{-4}$--$10^{-6}$) for deep CNNs and larger transformers, with the special case of small transformers (BERT-small) benefiting from large $ε$ ($ε=10$) due to its limited depth and unconstrained gradients.

Related papers

• EGGROLL, Unrolled: Understanding and Improving Low-Rank Evolution Strategies at Scale [9.562123195246393]
  EGGROLL makes evolution strategies practical for LLMs by replacing dense Gaussian weight perturbations with low-rank Gaussian products.<n>We analyze the error of its finite-population estimator.<n>LOO-ROLL, a leave-one-out estimator, preserves the finite-rank population field while replacing EGGROLL's two antithetic evaluations per direction by one.
  arXiv  Detail & Related papers  (2026-09-10T01:58:02Z)
• Reward-Aware Population Scaling of Evolutionary Strategies in LLM Fine-Tuning [0.0]
  Evolutionary Strategies for fine-tuning large language models is attractive because it is memory-efficient, parallel, and compatible with black-box or discrete rewards.<n>Yet its population-size conclusions conflict sharply: fine-tuning with cross-entropy reward succeeds with $N=1$, while binary-reward training often needs $N approx 30$.<n>We show this gap is largely about reward design and normalization, not population size.
  arXiv  Detail & Related papers  (2026-07-08T00:19:03Z)
• Beyond $\ell_2$-norm and $\ell_\infty$-norm: A Curvature-Inspired $\ell_p$-Norm Scheme for Deep Neural Networks [23.324496414025322]
  We propose a $ell_p$-norm scheme with a dynamical value of $p$ and incorporate it into a gradient descent (SGD) and momentum (SGDM)<n>We establish theoretical guarantees resulting in LPSGD and LPSGDM that achieve an (O-1/2) convergence rate for deep neural networks.
  arXiv  Detail & Related papers  (2026-06-01T11:07:35Z)
• Bug or Feature$^2$: Weight Drift, Activation Sparsity and Spikes [53.726365933748134]
  We analyze a negative weight drift induced by the interaction between standard losses and positively biased activation functions.<n>We characterize the sparsity-accuracy tradeoff across 79 configurations and identify a sharp accuracy cliff above $sim$70% activation sparsity.
  arXiv  Detail & Related papers  (2026-05-17T21:29:20Z)
• dynActivation: A Trainable Activation Family for Adaptive Nonlinearity [0.0]
  dynActivation variants tend to linearize deep layers while maintaining high performance.<n>A new proposed dynActGLU-variant achieves a $10.3%$ relative perplexity reduction over SwiGLU at 5620 steps (4.047 vs. 4.514), though the gap vanishes at 34300 steps.
  arXiv  Detail & Related papers  (2026-03-23T16:18:28Z)
• Evolution Strategies at the Hyperscale [57.75314521465674]
  We introduce EGGROLL, an evolution strategies (ES) algorithm designed to scale backprop-free optimization to large population sizes.<n>ES is a set of powerful blackbox optimisation methods that can handle non-differentiable or noisy objectives.<n>EGGROLL overcomes these bottlenecks by generating random matrices $Ain mathbbRmtimes r, Bin mathbbRntimes r$ with $rll min(m,n)$ to form a low-rank matrix perturbation $A Btop$
  arXiv  Detail & Related papers  (2025-11-20T18:56:05Z)
• Global Convergence of Gradient EM for Over-Parameterized Gaussian Mixtures [53.51230405648361]
  We study the dynamics of gradient EM and employ tensor decomposition to characterize the geometric landscape of the likelihood loss.<n>This is the first global convergence and recovery result for EM or gradient EM beyond the special case of $m=2$.
  arXiv  Detail & Related papers  (2025-06-06T23:32:38Z)
• FedSVD: Adaptive Orthogonalization for Private Federated Learning with LoRA [68.44043212834204]
  Low-Rank Adaptation (LoRA) is widely used for efficient fine-tuning of language models in learning (FL)<n>Low-Rank Adaptation (LoRA) is widely used for efficient fine-tuning of language models in learning (FL)
  arXiv  Detail & Related papers  (2025-05-19T07:32:56Z)
• Approximation Rates for Shallow ReLU$^k$ Neural Networks on Sobolev Spaces via the Radon Transform [12.171849953316192]
  We consider the problem of how efficiently shallow neural networks with the ReLU$k$ activation function can approximate functions from Sobolev spaces.<n>We provide a simple proof of nearly optimal approximation rates in a variety of cases, including when $qleq p$, $pgeq 2$, and $s leq k + (d+1)/2$.
  arXiv  Detail & Related papers  (2024-08-20T16:43:45Z)
• Generalization and Stability of Interpolating Neural Networks with Minimal Width [37.908159361149835]
  We investigate the generalization and optimization of shallow neural-networks trained by gradient in the interpolating regime. We prove the training loss number minimizations $m=Omega(log4 (n))$ neurons and neurons $Tapprox n$. With $m=Omega(log4 (n))$ neurons and $Tapprox n$, we bound the test loss training by $tildeO (1/)$.
  arXiv  Detail & Related papers  (2023-02-18T05:06:15Z)
• Bounding the Width of Neural Networks via Coupled Initialization -- A Worst Case Analysis [121.9821494461427]
  We show how to significantly reduce the number of neurons required for two-layer ReLU networks. We also prove new lower bounds that improve upon prior work, and that under certain assumptions, are best possible.
  arXiv  Detail & Related papers  (2022-06-26T06:51:31Z)
• Fine-Grained Gap-Dependent Bounds for Tabular MDPs via Adaptive Multi-Step Bootstrap [84.66885506098724]
  This paper presents a new model-free algorithm for episodic finite-horizon Markov Decision Processes (MDP), Adaptive Multi-step Bootstrap (AMB) We show AMB achieves a gap-dependent regret bound that only scales with the sum of the inverse of the sub-optimality gaps. We also show AMB suffers an additional $frac|Z_mul|Delta_min$ regret, where $Z_mul$ is the set of state-action pairs $(s,a)$'s satisfying $a$ is a non-unique optimal action for
  arXiv  Detail & Related papers  (2021-02-09T07:46:34Z)
• Least $k$th-Order and R\'{e}nyi Generative Adversarial Networks [12.13405065406781]
  Experimental results indicate that the proposed loss functions, applied to the MNIST and CelebA datasets, confer performance benefits by virtue of the extra degrees of freedom provided by the parameters $k$ and $alpha$, respectively. While it was applied to GANs in this study, the proposed approach is generic and can be used in other applications of information theory to deep learning, e.g., the issues of fairness or privacy in artificial intelligence.
  arXiv  Detail & Related papers  (2020-06-03T18:44:05Z)

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.