Litcius/Paper detail

PWLU: Learning Specialized Activation Functions With the Piecewise Linear Unit

Zezhou Zhu, Yucong Zhou, Yuan Dong, Zhao Zhong

2023IEEE Transactions on Pattern Analysis and Machine Intelligence13 citationsDOI

Abstract

The choice of activation functions is crucial to deep neural networks. ReLU is a popular hand-designed activation function. Swish, the automatically searched activation function, outperforms ReLU on many challenging datasets. However, the search method has two main drawbacks. First, the tree-based search space is highly discrete and restricted, which is difficult to search. Second, the sample-based search method is inefficient in finding specialized activation functions for each dataset or neural architecture. To overcome these drawbacks, we propose a new activation function called Piecewise Linear Unit (PWLU), incorporating a carefully designed formulation and learning method. PWLU can learn specialized activation functions for different models, layers, or channels. Besides, we propose a non-uniform version of PWLU, which maintains sufficient flexibility but requires fewer intervals and parameters. Additionally, we generalize PWLU to three-dimensional space to define a piecewise linear surface named 2D-PWLU, which can be treated as a non-linear binary operator. Experimental results show that PWLU achieves SOTA performance on various tasks and models, and 2D-PWLU is better than element-wise addition when aggregating features from different branches. The proposed PWLU and its variation are easy to implement and efficient for inference, which can be widely applied in real-world applications.

Topics & Concepts

Activation functionComputer sciencePiecewise linear functionArtificial intelligenceArtificial neural networkPiecewiseTree traversalInferenceFlexibility (engineering)Tree (set theory)Function (biology)AlgorithmPattern recognition (psychology)MathematicsGeometryBiologyStatisticsEvolutionary biologyMathematical analysisAdvanced Neural Network ApplicationsMachine Learning and Data ClassificationDomain Adaptation and Few-Shot Learning
PWLU: Learning Specialized Activation Functions With the Piecewise Linear Unit | Litcius