paper-with-me

홈 › Papers

Periodic Activation Functions Induce Stationarity

2021-10-26 · NeurIPS 2021 12 · Lassi Meronen, Martin Trapp, Arno Solin

Neural network models are known to reinforce hidden data biases, making them unreliable and difficult to interpret. We seek to build models that `know what they do not know' by introducing inductive biases in the function space. We show that periodic activation functions in Bayesian neural networks establish a connection between the prior on the network weights and translation-invariant, stationary Gaussian process priors. Furthermore, we show that this link goes beyond sinusoidal (Fourier) activations by also covering triangular wave and periodic ReLU activation functions. In a series of experiments, we show that periodic activation functions obtain comparable performance for in-domain data and capture sensitivity to perturbed inputs in deep neural networks for out-of-domain detection.

📄 PDF Abstract BibTeX arXiv:2110.13572

Code (2)

aaltoml/periodicbnn 공식 구현 pytorch
aaltoml/spatio-temporal-gps jax

Tasks

SensitivityTranslation

Methods 이 논문이 사용한 방법론

Gaussian Process Gaussian Processes are non-parametric models for approximating functions. They rely upon a measure of similarity between points (the kernel function) to predict the value for…

Similar Papers 제목 키워드 기반

Stationary Activations for Uncertainty Calibration in Deep Learning

2020-10-19 · NeurIPS 2020 12 · Lassi Meronen, Christabella Irwanto, Arno Solin

We introduce a new family of non-linear neural network activation functions that mimic the properties induced by the widely-used Mat\'ern family of kernels in Gaussian process (GP) models. This class spans a range of loc…

Deep LearningGeneral Classification

Extrapolation of Periodic Functions Using Binary Encoding of Continuous Numerical Values

2025-12-11 · Brian P. Powell, Jordan A. Caraballo-Vega, Mark L. Carroll, Thomas Maxwell 외 arxiv

We report the discovery that binary encoding allows neural networks to extrapolate periodic functions beyond their training bounds. We introduce Normalized Base-2 Encoding (NB2E) as a method for encoding continuous numer…

Frequency and Generalisation of Periodic Activation Functions in Reinforcement Learning

2024-07-09 · Augustine N. Mavor-Parker, Matthew J. Sargent, Caswell Barry, Lewis Griffin 외

Periodic activation functions, often referred to as learned Fourier features have been widely demonstrated to improve sample efficiency and stability in a variety of deep RL algorithms. Potentially incompatible hypothese…

reinforcement-learning

Neural Networks Fail to Learn Periodic Functions and How to Fix It

2020-06-15 · NeurIPS 2020 12 · Liu Ziyin, Tilman Hartwig, Masahito Ueda

Previous literature offers limited clues on how to learn a periodic function using modern neural networks. We start with a study of the extrapolation properties of neural networks; we prove and demonstrate experimentally…

Inductive Bias

FINER++: Building a Family of Variable-periodic Functions for Activating Implicit Neural Representation

2024-07-28 · Hao Zhu, Zhen Liu, Qi Zhang, Jingde Fu 외

Implicit Neural Representation (INR), which utilizes a neural network to map coordinate inputs to corresponding attributes, is causing a revolution in the field of signal processing. However, current INR techniques suffe…