-
NetReAct: Interactive Learning for Network Summarization
Authors:
Sorour E. Amiri,
Bijaya Adhikari,
John Wenskovitch,
Alexander Rodriguez,
Michelle Dowling,
Chris North,
B. Aditya Prakash
Abstract:
Generating useful network summaries is a challenging and important problem with several applications like sensemaking, visualization, and compression. However, most of the current work in this space do not take human feedback into account while generating summaries. Consider an intelligence analysis scenario, where the analyst is exploring a similarity network between documents. The analyst can ex…
▽ More
Generating useful network summaries is a challenging and important problem with several applications like sensemaking, visualization, and compression. However, most of the current work in this space do not take human feedback into account while generating summaries. Consider an intelligence analysis scenario, where the analyst is exploring a similarity network between documents. The analyst can express her agreement/disagreement with the visualization of the network summary via iterative feedback, e.g. closing or moving documents ("nodes") together. How can we use this feedback to improve the network summary quality? In this paper, we present NetReAct, a novel interactive network summarization algorithm which supports the visualization of networks induced by text corpora to perform sensemaking. NetReAct incorporates human feedback with reinforcement learning to summarize and visualize document networks. Using scenarios from two datasets, we show how NetReAct is successful in generating high-quality summaries and visualizations that reveal hidden patterns better than other non-trivial baselines.
△ Less
Submitted 21 December, 2020;
originally announced December 2020.
-
Investigation of Warrior Robots Behavior by Using Evolutionary Algorithms
Authors:
Shahriar Sharifi Borojerdi,
Mehdi Karimi,
Ehsan Amiri
Abstract:
In this study, we review robots behavior especially warrior robots by using evolutionary algorithms. This kind of algorithms is inspired by nature that causes robots behaviors get resemble to collective behavior. Collective behavior of creatures such as bees was shown that do some functions which depended on interaction and cooperation would need to a well-organized system so that all creatures wi…
▽ More
In this study, we review robots behavior especially warrior robots by using evolutionary algorithms. This kind of algorithms is inspired by nature that causes robots behaviors get resemble to collective behavior. Collective behavior of creatures such as bees was shown that do some functions which depended on interaction and cooperation would need to a well-organized system so that all creatures within it carry out their duty, very well. For robots which do not have any intelligence, we can define an algorithm and show the results by a simple simulation.
△ Less
Submitted 18 November, 2020;
originally announced November 2020.
-
Orientation-boosted Voxel Nets for 3D Object Recognition
Authors:
Nima Sedaghat,
Mohammadreza Zolfaghari,
Ehsan Amiri,
Thomas Brox
Abstract:
Recent work has shown good recognition results in 3D object recognition using 3D convolutional networks. In this paper, we show that the object orientation plays an important role in 3D recognition. More specifically, we argue that objects induce different features in the network under rotation. Thus, we approach the category-level classification task as a multi-task problem, in which the network…
▽ More
Recent work has shown good recognition results in 3D object recognition using 3D convolutional networks. In this paper, we show that the object orientation plays an important role in 3D recognition. More specifically, we argue that objects induce different features in the network under rotation. Thus, we approach the category-level classification task as a multi-task problem, in which the network is trained to predict the pose of the object in addition to the class label as a parallel task. We show that this yields significant improvements in the classification results. We test our suggested architecture on several datasets representing various 3D data sources: LiDAR data, CAD models, and RGB-D images. We report state-of-the-art results on classification as well as significant improvements in precision and speed over the baseline on 3D detection.
△ Less
Submitted 19 October, 2017; v1 submitted 12 April, 2016;
originally announced April 2016.
-
A revised Moore bound for mixed graphs
Authors:
Dominique Buset,
Mourad El Amiri,
Grahame Erskine,
Hebert Pérez-Rosés,
Mirka Miller
Abstract:
The degree-diameter problem seeks to find the maximum possible order of a graph with a given (maximum) degree and diameter. It is known that graphs attaining the maximum possible value (the Moore bound) are extremely rare, but much activity is focused on finding new examples of graphs or families of graph with orders approaching the bound as closely as possible.
There has been recent interest in…
▽ More
The degree-diameter problem seeks to find the maximum possible order of a graph with a given (maximum) degree and diameter. It is known that graphs attaining the maximum possible value (the Moore bound) are extremely rare, but much activity is focused on finding new examples of graphs or families of graph with orders approaching the bound as closely as possible.
There has been recent interest in this problem as it applies to mixed graphs, in which we allow some of the edges to be undirected and some directed. A 2008 paper of Nguyen and Miller derived an upper bound on the possible number of vertices of such graphs. We show that for diameters larger than three, this bound can be reduced and we present a corrected Moore bound for mixed graphs, valid for all diameters and for all combinations of undirected and directed degrees.
△ Less
Submitted 2 May, 2016; v1 submitted 11 August, 2015;
originally announced August 2015.
