MorphoActivation: Generalizing ReLU Activation Function by Mathematical Morphology

被引：2

作者：

Velasco-Forero, Santiago ^{[1
]}

Angulo, Jesus ^{[1
]}

机构：

[1] PSL Univ, Ctr Math Morphol CMM, MINES Paris, Paris, France

来源：

DISCRETE GEOMETRY AND MATHEMATICAL MORPHOLOGY, DGMM 2022 | 2022年 / 13493卷

关键词：

Matheron's representation theory; Activation function; Mathematical morphology; Deep learning; REPRESENTATIONS;

D O I：

10.1007/978-3-031-19897-7_35

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

This paper analyses both nonlinear activation functions and spatial max-pooling for Deep Convolutional Neural Networks (DCNNs) by means of the algebraic basis of mathematical morphology. Additionally, a general family of activation functions is proposed by considering both max-pooling and nonlinear operators in the context of morphological representations. Experimental section validates the goodness of our approach on classical benchmarks for supervised learning by DCNN.

引用

页码：449 / 461

页数：13

共 50 条

[1] An Empirical Study on Generalizations of the ReLU Activation Function
Banerjee, Chaity
Mukherjee, Tathagata
Pasiliao, Eduardo, Jr.
PROCEEDINGS OF THE 2019 ANNUAL ACM SOUTHEAST CONFERENCE (ACMSE 2019), 2019, : 164 - 167
[2] The Multi-phase ReLU Activation Function
Banerjee, Chaity
Mukherjee, Tathagata
Pasiliao, Eduardo, Jr.
ACMSE 2020: PROCEEDINGS OF THE 2020 ACM SOUTHEAST CONFERENCE, 2020, : 239 - 242
[3] RMAF: Relu-Memristor-Like Activation Function for Deep Learning
Yu, Yongbin
Adu, Kwabena
Tashi, Nyima
Anokye, Patrick
Wang, Xiangxiang
Ayidzoe, Mighty Abra
IEEE ACCESS, 2020, 8 : 72727 - 72741
[4] Learnable Leaky ReLU (LeLeLU): An Alternative Accuracy-Optimized Activation Function
Maniatopoulos, Andreas
Mitianoudis, Nikolaos
INFORMATION, 2021, 12 (12)
[5] Hypercomplex Mathematical Morphology
Angulo, Jesus
JOURNAL OF MATHEMATICAL IMAGING AND VISION, 2011, 41 (1-2) : 86 - 108
[6] Optimization of ReLU Activation Function for Deep-Learning-based Breast Cancer Classification on Mammogram Images
Razali, Noor Fadzilah
Isa, Iza Sazanita
Sulaiman, Siti Noraini
Osman, Muhammad Khusairi
Karim, Noor Khairiah A.
Nordin, Siti Aminah
2024 IEEE INTERNATIONAL CONFERENCE ON AUTOMATIC CONTROL AND INTELLIGENT SYSTEMS, I2CACIS 2024, 2024, : 267 - 272
[7] Global-connected network with generalized ReLU activation
Chen, Zhi
Ho, Pin-Han
PATTERN RECOGNITION, 2019, 96
[8] Deep Learning-Based Beamforming for Millimeter-Wave Systems Using Parametric ReLU Activation Function
Alshimaa H. Ismail
Tarek Abed Soliman
Mohamed Rihan
Moawad I. Dessouky
Wireless Personal Communications, 2023, 129 : 825 - 836
[9] Deep Learning-Based Beamforming for Millimeter-Wave Systems Using Parametric ReLU Activation Function
Ismail, Alshimaa H. H.
Soliman, Tarek Abed
Rihan, Mohamed
Dessouky, Moawad I. I.
WIRELESS PERSONAL COMMUNICATIONS, 2023, 129 (02) : 825 - 836
[10] ONLINE FUNCTION MINIMIZATION WITH CONVEX RANDOM RELU EXPANSIONS
Bliek, Laurens
Verhaegen, Michel
Wahls, Sander
2017 IEEE 27TH INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING, 2017,

← 1 2 3 4 5 →