[Zurück]

H. Eidenberger, B. Klauninger, M. Unger:
"Machine Learning with Dual Process Models";
Vortrag: 5th International Conference on Pattern Recognition Applications and Methods, Rom; 24.02.2016 - 26.02.2016; in: "Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods", (2016), ISBN: 989-758-173-1; S. 148 - 153.

Similarity measurement processes are a core part of most machine learning algorithms. Traditional approaches
focus on either taxonomic or thematic thinking. Psychological research suggests that a combination of both
is needed to model human-like similarity perception adequately. Such a combination is called a Similarity
Dual Process Model (DPM). This paper describes how to construct DPMs as a linear combination of existing
measures of similarity and distance. We use generalisation functions to convert distance into similarity. DPMs
are similar to kernel functions. Thus, they can be integrated into any machine learning algorithm that uses
kernel functions.Clearly, not all DPMs that can be formulated work equally well. Therefore we test classification
performance in a real-world task: the detection of pedestrians in images. We assume that DPMs are only
viable if they yield better classifiers than their constituting parts. In our experiments, we found DPM kernels
that matched the performance of conventional ones for our data set. Eventually, we provide a construction kit
to build such kernels to encourage further experiments in other application domains of machine learning.

Similarity measurement processes are a core part of most machine learning algorithms. Traditional approaches
focus on either taxonomic or thematic thinking. Psychological research suggests that a combination of both
is needed to model human-like similarity perception adequately. Such a combination is called a Similarity
Dual Process Model (DPM). This paper describes how to construct DPMs as a linear combination of existing
measures of similarity and distance. We use generalisation functions to convert distance into similarity. DPMs
are similar to kernel functions. Thus, they can be integrated into any machine learning algorithm that uses
kernel functions.Clearly, not all DPMs that can be formulated work equally well. Therefore we test classification
performance in a real-world task: the detection of pedestrians in images. We assume that DPMs are only
viable if they yield better classifiers than their constituting parts. In our experiments, we found DPM kernels
that matched the performance of conventional ones for our data set. Eventually, we provide a construction kit
to build such kernels to encourage further experiments in other application domains of machine learning.

http://dx.doi.org/10.5220/0005655901480153