Yangdong Ye

miSFM: On combination of mutual information and social force model towards simulating crowd evacuation

Abstract In this paper we propose a novel technique termed miSFM for the simulation of crowd evacuation. miSFM take merits of both Mutual Information (MI) and Social Force Model (SFM). More specifically, MI of interacting agents is adopted to determine the level of order within a crowd during an evacuation. In such a way, SFM can be improved by adapting the forces involved at microscopic level between mutually interacting agents. The key innovation lies in highlighting how the dynamic adjustment of SFM parameters reveals much more realistic crowd movements for the evacuation simulation. Extensive experiments over several alternative and state-of-the-art works demonstrate the advantages of the proposed algorithm.

Coherent Motion Detection with Collective Density Clustering

Detecting coherent motion is significant for analysing the crowd motion in video applications. In this study, we propose the Collective Density Clustering(CDC) approach to recognize both local and global coherent motion having arbitrary shapes and varying densities. Firstly, the collective density is defined to reveal the underlying patterns with varying levels of density. Based on collective density, the collective clustering algorithm is further presented to recognize the local consistency, where density-based clustering is more adaptive to recognize clusters with arbitrary shapes. This algorithm has salient properties including single step of clustering process, automatical decision of clustering number and accurate identification of outliers. Finally, the collective merging algorithm is introduced to fully characterize the global consistency. Experiments on diverse crowd scenes, including pedestrians, traffic and bacterial colony, demonstrate the effectiveness for coherent motion detection. The comparisons show that our approach outperforms state-of-the-art coherent detection techniques.

AA-FVDM: An accident-avoidance full velocity difference model for animating realistic street-level traffic in rural scenes

Most of existing traffic simulation efforts focus on urban regions with a coarse two‐dimensional representation; relatively few studies have been conducted to simulate realistic three‐dimensional traffic flows on a large, complex road web in rural scenes. In this paper, we present a novel agent‐based approach called accident‐avoidance full velocity difference model (abbreviated as AA‐FVDM) to simulate realistic street‐level rural traffics, on top of the existing FVDM. The main distinction between FVDM and AA‐FVDM is that FVDM cannot handle a critical real‐world traffic problem while AA‐FVDM settles this problem and retains the essence of FVDM. We also design a novel scheme to animate the lane‐changing maneuvering process (in particular, the execution course). Through numerous simulations, we demonstrate that besides addressing a previously unaddressed real‐world traffic problem, our AA‐FVDM method efficiently (in real time) simulates large‐scale traffic flows (tens of thousands of vehicles) with realistic, smooth effects. Furthermore, we validate our method using real‐world traffic data, and the validation results show that our method measurably outperforms state‐of‐the‐art traffic simulation methods.Copyright

Moving-Target Pursuit Algorithm Using Improved Tracking Strategy

Pursuing a moving target in modern computer games presents several challenges to situated agents, including real-time response, large-scale search space, severely limited computation resources, incomplete environmental knowledge, adversarial escaping strategy, and outsmarting the opponent. In this paper, we propose a novel tracking automatic optimization moving-target pursuit (TAO-MTP) algorithm employing improved tracking strategy to effectively address all challenges above for the problem involving single hunter and single prey. TAO-MTP uses a queue to store preys trajectory, and simultaneously runs real-time adaptive A* (RTAA*) repeatedly to approach the optimal position updated periodically in the trajectory within limited steps, which makes the overall pursuit cost smallest. In the process, the hunter speculatively moves to any position explored in the trajectory, not necessarily the optimal position, to speed up convergence, and then directly moves along the trajectory to pursue the prey. Moreover, automatic optimization methods, such as reducing trajectory storage and optimizing pursuit path, are used to further enhance its performance. As long as the hunters moving speed is faster than that of the prey, and its sense scope is large enough, it will eventually capture the prey. Experiments using commercial game maps show that TAO-MTP is independent of adversarial escaping strategy, and outperforms all the classic and state-of-the-art moving-target pursuit algorithms such as extended moving-target search (eMTS), path refinement moving-target search (PR MTS), moving-target adaptive A* (MTAA*), and generalized adaptive A* (GAA*).

Transfer active learning

Active learning traditionally assumes that labeled and unlabeled samples are subject to the same distributions and the goal of an active learner is to label the most informative unlabeled samples. In reality, situations may exist that we may not have unlabeled samples from the same domain as the labeled samples (i.e. target domain), whereas samples from auxiliary domains might be available. Under such situations, an interesting question is whether an active learner can actively label samples from auxiliary domains to benefit the target domain. In this paper, we propose a transfer active learning method, namely Transfer Active SVM (TrAcSVM), which uses a limited number of target instances to iteratively discover and label informative auxiliary instances. TrAcSVM employs an extended sigmoid function as instance weight updating approach to adjust the models for prediction of (newly arrived) target data. Experimental results on real-world data sets demonstrate that TrAcSVM obtains better efficiency and prediction accuracy than its peers.

Unsupervised object category discovery via information bottleneck method

We present a novel approach to automatically discover object categories from a collection of unlabeled images. This is achieved by the Information Bottleneck method, which finds the optimal partitioning of the image collection by maximally preserving the relevant information with respect to the latent semantic residing in the image contents. In this method, the images are modeled by the Bag-of-Words representation, which naturally transforms each image into a visual document composed of visual words. Then the sIB algorithm is adopted to learn the object patterns by maximizing the semantic correlations between the images and their constructive visual words. Extensive experimental results on 15 benchmark image datasets show that the Information Bottleneck method is a promising technique for discovering the hidden semantic of images, and is superior to the state-of-the-art unsupervised object category discovery methods.

Fuzzy peak hour for urban road traffic network

Traffic congestion is now nearly ubiquitous in many urban areas and frequently occurs during rush hour periods. Rush hour avoidance is an effective way to ease traffic congestion. It is significant to calculate the rush hour for alleviating traffic congestion. This paper provides a method to calculate the fuzzy peak hour of the urban traffic network considering the flow, speed and occupancy. The process of calculation is based on betweenness centrality of network theory, optimal separation method, time period weighting, probability–possibility transformations and trapezoidal approximations of fuzzy numbers. The fuzzy peak hour of the urban road traffic network (URTN) is a trapezoidal fuzzy number [m1, m2, m3, m4]. It helps us (i) to confirm a more detailed traffic condition at each moment, (ii) to distinguish the five traffic states of the traffic network in one day, (iii) to analyze the characteristic of appearance and disappearance processes of the each traffic state and (iv) to find out the time pattern of residents travel in one city.

Inverse matrix-free incremental proximal support vector machine

Traditional Support Vector Machines (SVMs) based learners are commonly regarded as strong classifiers for many learning tasks. Their efficiency for large-scale high dimensional data, however, has shown to be unsatisfactory. Consequently, many alternative SVM solutions exist for large-scale and/or high dimensional data. Among them, proximal support vector machine (PSVM) is a simple but effective SVM classifier. Its incremental version (ISVM) is also available for large-scale data. Nevertheless, the computational efficiency of the ISVM for high dimensional data still needs to be improved, mainly because it requires explicit matrix inversion for updating the decision model. To solve this problem, we propose, in this paper, an inverse matrix-free incremental PSVM (IMISVM) with the following two characteristics. Firstly, IMISVM avoids explicit matrix inversion and hence derives simple formulas for updating model parameters. Secondly, IMISVM achieves faster convergence speed than ISVM. Experimental results on synthetic and real-world data sets confirm that the proposed incremental classifier outperforms ISVM.

Instance-based ensemble pruning for imbalanced learning

Class-imbalance is very common in real world. However, traditional state-of-the-art classifiers do not work well on imbalanced data sets for imbalanced class distribution. This paper considers imbalance learning from the viewpoint of ensemble pruning, and proposes a novel approach called IBEP (Instance-Based Ensemble Pruning) to improve classifier’s performance on these data sets. Unlike traditional approaches which consider imbalance problem in training stage, IBEP focuses on the problem in prediction stage. Given an unlabeled instance, IBEP tries to search for the k nearest neighbors as the corresponding pruning set and adopts ensemble pruning strategy to select a subset of ensemble members to form sub-ensemble based on the pruning set to predict the instance. In this way, IBEP pays more attention to rare class and achieves better performance on imbalanced data set. Besides, two widely used sampling techniques, under-sampling and SMOTE, are skillfully combined with IBEP to further improve its performance. Experimental results on 14 data sets show that IBEP performs significantly better than many state-of-the-art classification methods on all metrics used in this paper including recall, f -measure and g-mean.

Analysis of temporal uncertainty of trains converging based on fuzzy time petri nets

The paper defines a fuzzy time Petri net (FTPN) which adopts four fuzzy set theoretic functions of time called fuzzy timestamp, fuzzy enabling time, fuzzy occurrence time and fuzzy delay, to deal with temporal uncertainty of train group operation and we also present different firing strategies for the net to give prominence to key events. The application instance shows that the method based on FTPN can efficiently analyze trains converging time, the possibility of converging and train terminal time in adjustment of train operation plan. Compared with time interval method, this method has some outstanding characteristics such as accurate analysis, simple computation, system simplifying and convenience for system integrating.