Convolutional Neural Networks:: Key Components and Architectures

Manfred M. Fischer

No 1, Working Papers in Regional Science from WU Vienna University of Economics and Business

Abstract: Convolutional Neural Networks (CNNs) are a specialized class of deep neural networks tailored for processing high-dimensional data, excelling in tasks like image classification, object detection, and facial recognition. Their architecture is built on convolutional layers interspersed with pooling layers, which efficiently extract hierarchical features while reducing computational complexity. Convolutional layers utilize filters to detect patterns such as edges or textures, employing shared weights to enhance translational invariance and reduce the number of parameters. Pooling layers further simplify feature maps by downsampling, preserving critical information while minimizing spatial dimensions. Prominent architectures, including LeNet, AlexNet, VGGNet, and ResNet, have set benchmarks in image recognition and inspired advancements in deep learning. The careful tuning of hyperparameters, such as filter size, stride, and padding, plays a pivotal role in optimizing performance, balancing accuracy with computational efficiency. CNNs continue to drive innovation, expanding their applications across diverse fields like natural language processing, speech recognition, and autonomous systems.

Keywords: Deep convolutional neural networks,; LeNet; AlexNet; VGGNet; ResNet (search for similar items in EconPapers)
Date: 2025
New Economics Papers: this item is included in nep-big and nep-cmp
References: Add references at CitEc
Citations:

Downloads: (external link)
https://research.wu.ac.at/en/publications/2076fa6b-8225-433c-b3e6-c96de2e46735 original version (application/pdf)
Our link check indicates that this URL is bad, the error code is: 403 Forbidden

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:wiw:wus046:71119092

Access Statistics for this paper

More papers in Working Papers in Regional Science from WU Vienna University of Economics and Business Welthandelsplatz 1, 1020 Vienna, Austria.
Bibliographic data for series maintained by WU Library ().