Pairwise comparison

About: Pairwise comparison is a research topic. Over the lifetime, 6804 publications have been published within this topic receiving 174081 citations.

A Pairwise Attentive Adversarial Spatiotemporal Network for Cross-Domain Few-Shot Action Recognition-R2

Abstract: - Action recognition is a popular research topic in the computer vision and machine learning domains. Although many action recognition methods have been proposed, only a few researchers have focused on cross-domain few-shot action recognition, which must often be performed in real security surveillance. Since the problems of action recognition, domain adaptation, and few-shot learning need to be simultaneously solved, the cross-domain few-shot action recognition task is a challenging problem. To solve these issues, in this work, we develop a novel end-to-end pairwise attentive adversarial spatiotemporal network ( PASTN ) to perform the cross-domain few-shot action recognition task, in which spatiotemporal information acquisition, few-shot learning, and video domain adaptation are realised in a unified framework. Specifically, the Resnet-50 network is selected as the backbone of the PASTN , and a 3D convolution block is embedded in the top layer of the 2D CNN (ResNet-50) to capture the spatiotemporal representations. Moreover, a novel attentive adversarial network architecture is designed to align the spatiotemporal dynamics actions with higher domain discrepancies. In addition, the pairwise margin discrimination loss is designed for the pairwise network architecture to improve the discrimination of the learned domain-invariant spatiotemporal feature. The results of extensive experiments performed on three public benchmarks of the cross-domain action recognition datasets, including SDAI Action I , SDAI Action II and UCF50-OlympicSport, demonstrate that the proposed PASTN can significantly outperform the state-of-the-art cross-domain action recognition methods in terms of both the accuracy and computational time. Even when only two labelled training samples per category are considered in the office1 scenario of the SDAI Action I dataset, the accuracy of the PASTN is improved by 6.1%, 10.9%, 16.8%, and 14% compared to that of the $TA^{3}N$ , TemporalPooling, I3D, and P3D methods, respectively.

The computational complexity of choice sets

Abstract: Social choice rules are often evaluated and compared by inquiring whether they fulfill certain desirable criteria such as the Condorcet criterion, which states that an alternative should always be chosen when more than half of the voters prefer it over any other alternative. Many of these criteria can be formulated in terms of choice sets that single out reasonable alternatives based on the preferences of the voters. In this paper, we consider choice sets whose definition merely relies on the pairwise majority relation. These sets include the Copeland set, the Smith set, the Schwartz set, von Neumann-Morgenstern stable sets (a concept originally introduced in the context of cooperative game theory), the Banks set, and the Slater set. We investigate the relationships between these sets and completely characterize their computational complexity which allows us to obtain hardness results for entire classes of social choice rules. In contrast to most existing work, we do not impose any restrictions on individual preferences, apart from the indifference relation being reflexive and symmetric. This assumption is motivated by the fact that many realistic types of preferences in computational contexts such as incomplete or quasi-transitive preferences may lead to general pairwise majority relations that need not be complete.

A note on the information content of a consistent pairwise comparison judgment matrix of an AHP decision maker

Abstract: A decision maker using the Analytic Hierarchy Process (AHP) could be consistent, and still provide no information in the resulting vector of priorities. An extreme example would be a pairwise comparison judgment matrix filled with 1s which is totally consistent under the various definitions of consistency, but has provided no information about the prioritization of alternatives resulting from the decision maker's judgments. In this paper, the quality of a consistent decision maker's judgments using the Analytic Hierarchy Process is placed in the context of the entropy of the resulting vector of priorities. Indeed, it is the purpose of this paper to provide a formal definition of this notion ofentropy of a priority vector, and to provide a framework for a quantitative measurement of the information content of consistent pairwise comparison judgment matrices of a decision maker who is using the Analytic Hierarchy Process. We will prove that the entropy of the vector of priorities for consistent matrices follows a normal distribution and discuss some general considerations of this result.

Metrics for Graph Comparison: A Practitioner's Guide

Abstract: Comparison of graph structure is a ubiquitous task in data analysis and machine learning, with diverse applications in fields such as neuroscience, cyber security, social network analysis, and bioinformatics, among others. Discovery and comparison of structures such as modular communities, rich clubs, hubs, and trees in data in these fields yields insight into the generative mechanisms and functional properties of the graph. Often, two graphs are compared via a pairwise distance measure, with a small distance indicating structural similarity and vice versa. Common choices include spectral distances (also known as $\lambda$ distances) and distances based on node affinities. However, there has of yet been no comparative study of the efficacy of these distance measures in discerning between common graph topologies and different structural scales. In this work, we compare commonly used graph metrics and distance measures, and demonstrate their ability to discern between common topological features found in both random graph models and empirical datasets. We put forward a multi-scale picture of graph structure, in which the effect of global and local structure upon the distance measures is considered. We make recommendations on the applicability of different distance measures to empirical graph data problem based on this multi-scale view. Finally, we introduce the Python library NetComp which implements the graph distances used in this work.