Preliminaries of Information Theory and Neural Networks

Conventional convolutional neural networks (CNNs) have achieved great success in image semantic segmentation. Existing methods mainly focus on learning pixel-wise labels from an image directly. In this paper, we advocate tackling the pixel-wise segmentation problem by considering the image-level classification labels. Theoretically, we analyze and discuss the effects of image-level labels on pixel-wise segmentation from the perspective of information theory. In practice, an end-to-end segmentation model is built by fusing the image-level and pixel-wise labeling networks. A generative network is included to reconstruct the input image and further boost the segmentation model training with an auxiliary loss. Extensive experimental results on benchmark dataset demonstrate the effectiveness of the proposed method, where good image-level labels can significantly improve the pixel-wise segmentation accuracy.

Download Full-text

An Analysis of Internal Representations for Two Artificial Neural Networks that Classify Musical Chords

Discoveries and Breakthroughs in Cognitive Informatics and Natural Intelligence ◽

10.4018/978-1-60566-902-1.ch026 ◽

2010 ◽

pp. 485-507

Author(s):

Vanessa Yaremchuk

Keyword(s):

Neural Networks ◽

Information Theory ◽

Probability Theory ◽

Artificial Neural Networks ◽

Intelligent Agents ◽

Natural World ◽

External Signal ◽

Cognitive Informatics ◽

Artificial Neural ◽

Internal Processing

Cognitive informatics is a field of research that is primarily concerned with the information processing of intelligent agents; it can be characterised in terms of an evolving notion of information (Wang, 2007). When it originated six decades ago, conventional accounts of information were concerned about using probability theory and statistics to measure the amount of information carried by an external signal. This, in turn, developed into the notion of modern informatics which studied information as “properties or attributes of the natural world that can be generally abstracted, quantitatively represented, and mentally processed” (Wang, 2007, p. iii). The current incarnation of cognitive informatics recognised that both information theory and modern informatics defined information in terms of factors that were external to brains, and has replaced this with an emphasis on exploring information as an internal property. This emphasis on the internal processing of information raises fundamental questions about how such information can be represented. One approach to answering such questions — and for proposing new representational accounts — would be to train a brain-like system to perform an intelligent task, and then to analyse its internal structure to determine the types of representations that the system had developed to perform this intelligent behaviour. The logic behind this approach is that when artificial neural networks Cognitive informatics is a field of research that is primarily concerned with the information processing of intelligent agents; it can be characterised in terms of an evolving notion of information (Wang, 2007). When it originated six decades ago, conventional accounts of information were concerned about using probability theory and statistics to measure the amount of information carried by an external signal. This, in turn, developed into the notion of modern informatics which studied information as “properties or attributes of the natural world that can be generally abstracted, quantitatively represented, and mentally processed” (Wang, 2007, p. iii). The current incarnation of cognitive informatics recognised that both information theory and modern informatics defined information in terms of factors that were external to brains, and has replaced this with an emphasis on exploring information as an internal property. This emphasis on the internal processing of information raises fundamental questions about how such information can be represented. One approach to answering such questions — and for proposing new representational accounts — would be to train a brain-like system to perform an intelligent task, and then to analyse its internal structure to determine the types of representations that the system had developed to perform this intelligent behaviour. The logic behind this approach is that when artificial neural networks

Download Full-text

Unsupervised information theory-based training algorithms for multilayer neural networks

[Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing ◽

10.1109/icassp.1992.225889 ◽

1992 ◽

Cited By ~ 3

Author(s):

G. Rigoll

Keyword(s):

Neural Networks ◽

Information Theory ◽

Training Algorithms ◽

Multilayer Neural Networks

Download Full-text