Mining Generalizable Activation Functions

• URL: http://arxiv.org/abs/2602.05688v1
• Date: Thu, 05 Feb 2026 14:13:40 GMT
• Title: Mining Generalizable Activation Functions
• Authors: Alex Vitvitskyi, Michael Boratko, Matej Grcic, Razvan Pascanu, Deep Shah, Petar Veličković,
• Abstract summary: We argue that evolutionary search provides a useful framework for finding new activation functions.<n>We show that relatively small scale synthetic datasets can be sufficient for AlphaEvolve to discover meaningful activations.
• Score: 24.370797575430174
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The choice of activation function is an active area of research, with different proposals aimed at improving optimization, while maintaining expressivity. Additionally, the activation function can significantly alter the implicit inductive bias of the architecture, controlling its non-linear behavior. In this paper, in line with previous work, we argue that evolutionary search provides a useful framework for finding new activation functions, while we also make two novel observations. The first is that modern pipelines, such as AlphaEvolve, which relies on frontier LLMs as a mutator operator, allows for a much wider and flexible search space; e.g., over all possible python functions within a certain FLOP budget, eliminating the need for manually constructed search spaces. In addition, these pipelines will be biased towards meaningful activation functions, given their ability to represent common knowledge, leading to a potentially more efficient search of the space. The second observation is that, through this framework, one can target not only performance improvements but also activation functions that encode particular inductive biases. This can be done by using performance on out-of-distribution data as a fitness function, reflecting the degree to which the architecture respects the inherent structure in the data in a manner independent of distribution shifts. We carry an empirical exploration of this proposal and show that relatively small scale synthetic datasets can be sufficient for AlphaEvolve to discover meaningful activations.

Related papers

• Efficient Search for Customized Activation Functions with Gradient Descent [42.20716255578699]
  Different activation functions work best for different deep learning models. We propose a fine-grained search cell that combines basic mathematical operations to model activation functions. Our approach enables the identification of specialized activations, leading to improved performance in every model we tried.
  arXiv  Detail & Related papers  (2024-08-13T11:27:31Z)
• Efficient Model-Free Exploration in Low-Rank MDPs [76.87340323826945]
  Low-Rank Markov Decision Processes offer a simple, yet expressive framework for RL with function approximation. Existing algorithms are either (1) computationally intractable, or (2) reliant upon restrictive statistical assumptions. We propose the first provably sample-efficient algorithm for exploration in Low-Rank MDPs.
  arXiv  Detail & Related papers  (2023-07-08T15:41:48Z)
• Efficient Activation Function Optimization through Surrogate Modeling [15.219959721479835]
  This paper aims to improve the state of the art through three steps. First, the benchmark Act-Bench-CNN, Act-Bench-ResNet, and Act-Bench-ViT were created by training convolutional, residual, and vision transformer architectures. Second, a characterization of the benchmark space was developed, leading to a new surrogate-based method for optimization.
  arXiv  Detail & Related papers  (2023-01-13T23:11:14Z)
• Offline Reinforcement Learning with Differentiable Function Approximation is Provably Efficient [65.08966446962845]
  offline reinforcement learning, which aims at optimizing decision-making strategies with historical data, has been extensively applied in real-life applications. We take a step by considering offline reinforcement learning with differentiable function class approximation (DFA) Most importantly, we show offline differentiable function approximation is provably efficient by analyzing the pessimistic fitted Q-learning algorithm.
  arXiv  Detail & Related papers  (2022-10-03T07:59:42Z)
• Transformers with Learnable Activation Functions [63.98696070245065]
  We use Rational Activation Function (RAF) to learn optimal activation functions during training according to input data. RAF opens a new research direction for analyzing and interpreting pre-trained models according to the learned activation functions.
  arXiv  Detail & Related papers  (2022-08-30T09:47:31Z)
• Stabilizing Q-learning with Linear Architectures for Provably Efficient Learning [53.17258888552998]
  This work proposes an exploration variant of the basic $Q$-learning protocol with linear function approximation. We show that the performance of the algorithm degrades very gracefully under a novel and more permissive notion of approximation error.
  arXiv  Detail & Related papers  (2022-06-01T23:26:51Z)
• Efficient Neural Network Analysis with Sum-of-Infeasibilities [64.31536828511021]
  Inspired by sum-of-infeasibilities methods in convex optimization, we propose a novel procedure for analyzing verification queries on networks with extensive branching functions. An extension to a canonical case-analysis-based complete search procedure can be achieved by replacing the convex procedure executed at each search state with DeepSoI.
  arXiv  Detail & Related papers  (2022-03-19T15:05:09Z)
• Learning specialized activation functions with the Piecewise Linear Unit [7.820667552233989]
  We propose a new activation function called Piecewise Linear Unit(PWLU), which incorporates a carefully designed formulation and learning method. It can learn specialized activation functions and achieves SOTA performance on large-scale datasets like ImageNet and COCO. PWLU is also easy to implement and efficient at inference, which can be widely applied in real-world applications.
  arXiv  Detail & Related papers  (2021-04-08T11:29:11Z)
• Learning Discrete Energy-based Models via Auxiliary-variable Local Exploration [130.89746032163106]
  We propose ALOE, a new algorithm for learning conditional and unconditional EBMs for discrete structured data. We show that the energy function and sampler can be trained efficiently via a new variational form of power iteration. We present an energy model guided fuzzer for software testing that achieves comparable performance to well engineered fuzzing engines like libfuzzer.
  arXiv  Detail & Related papers  (2020-11-10T19:31:29Z)
• Discovering Parametric Activation Functions [17.369163074697475]
  This paper proposes a technique for customizing activation functions automatically, resulting in reliable improvements in performance. Experiments with four different neural network architectures on the CIFAR-10 and CIFAR-100 image classification datasets show that this approach is effective.
  arXiv  Detail & Related papers  (2020-06-05T00:25:33Z)
• Evolutionary Optimization of Deep Learning Activation Functions [15.628118691027328]
  We show that evolutionary algorithms can discover novel activation functions that outperform the Rectified Linear Unit (ReLU) replacing ReLU with evolved activation functions results in statistically significant increases in network accuracy. These novel activation functions are shown to generalize, achieving high performance across tasks.
  arXiv  Detail & Related papers  (2020-02-17T19:54:26Z)

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.