The design of activation functions is a growing research area in the field of neural networks. In particular, instead of using fixed point-wise functions (e.g., the rectified linear unit), several authors have proposed ways of learning these functions directly from the data in a non-parametric fashion. In this paper we focus on the kernel activation function (KAF), a recently proposed framework wherein each function is modeled as a one-dimensional kernel model, whose weights are adapted through standard backpropagation-based optimization. One drawback of KAFs is the need to select a single kernel function and its eventual hyper-parameters. To partially overcome this problem, we motivate an extension of the KAF model, in which multiple kernels are linearly combined at every neuron, inspired by the literature on multiple kernel learning. We provide an application of the resulting multi-KAF on a realistic use case, specifically handwritten Latin OCR, on a large dataset collected in the context of the ‘In Codice Ratio’ project. Results show that multi-KAFs can improve the accuracy of the convolutional networks previously developed for the task, with faster convergence, even with a smaller number of overall parameters.
Multikernel Activation Functions: Formulation and a Case Study / Scardapane, Simone; Nieddu, Elena; Firmani, Donatella; Merialdo, Paolo. - (2020), pp. 320-329. (Intervento presentato al convegno 2019 INNS Big Data and Deep Learning conference tenutosi a Sestri Levante; Italy) [10.1007/978-3-030-16841-4_33].
Multikernel Activation Functions: Formulation and a Case Study
Scardapane, Simone
;Nieddu, Elena;Firmani, Donatella;
2020
Abstract
The design of activation functions is a growing research area in the field of neural networks. In particular, instead of using fixed point-wise functions (e.g., the rectified linear unit), several authors have proposed ways of learning these functions directly from the data in a non-parametric fashion. In this paper we focus on the kernel activation function (KAF), a recently proposed framework wherein each function is modeled as a one-dimensional kernel model, whose weights are adapted through standard backpropagation-based optimization. One drawback of KAFs is the need to select a single kernel function and its eventual hyper-parameters. To partially overcome this problem, we motivate an extension of the KAF model, in which multiple kernels are linearly combined at every neuron, inspired by the literature on multiple kernel learning. We provide an application of the resulting multi-KAF on a realistic use case, specifically handwritten Latin OCR, on a large dataset collected in the context of the ‘In Codice Ratio’ project. Results show that multi-KAFs can improve the accuracy of the convolutional networks previously developed for the task, with faster convergence, even with a smaller number of overall parameters.File | Dimensione | Formato | |
---|---|---|---|
Scardapane_Multikernel_2020.pdf
solo gestori archivio
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
708.44 kB
Formato
Adobe PDF
|
708.44 kB | Adobe PDF | Contatta l'autore |
Scardapane_preprint_Multikernel_2019.pdf
accesso aperto
Note: https://doi.org/10.1007/978-3-030-16841-4_33
Tipologia:
Documento in Pre-print (manoscritto inviato all'editore, precedente alla peer review)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
1.88 MB
Formato
Adobe PDF
|
1.88 MB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.