An Uncertainty-based Neural Network for Explainable Trajectory Segmentation

2022 ◽  
Vol 13 (1) ◽  
pp. 1-18
Author(s):  
Xin Bi ◽  
Chao Zhang ◽  
Fangtong Wang ◽  
Zhixun Liu ◽  
Xiangguo Zhao ◽  
...  
Keyword(s):  
Real World ◽  
Feature Learning ◽  
Learning Performance ◽  
Bed Net ◽  
Time Series Segmentation ◽  
Real World Applications ◽  
Real World Datasets ◽  
High Level ◽  
High Level Feature ◽  
Trajectory Segmentation

As a variant task of time-series segmentation, trajectory segmentation is a key task in the applications of transportation pattern recognition and traffic analysis. However, segmenting trajectory is faced with challenges of implicit patterns and sparse results. Although deep neural networks have tremendous advantages in terms of high-level feature learning performance, deploying as a blackbox seriously limits the real-world applications. Providing explainable segmentations has significance for result evaluation and decision making. Thus, in this article, we address trajectory segmentation by proposing a Bayesian Encoder-Decoder Network (BED-Net) to provide accurate detection with explainability and references for the following active-learning procedures. BED-Net consists of a segmentation module based on Monte Carlo dropout and an explanation module based on uncertainty learning that provides results evaluation and visualization. Experimental results on both benchmark and real-world datasets indicate that BED-Net outperforms the rival methods and offers excellent explainability in the applications of trajectory segmentation.

scholarly journals Domain Conditioned Adaptation Network

2020 ◽  
Vol 34 (07) ◽  
pp. 11386-11393 ◽  
Author(s):  
Shuang Li ◽  
Chi Liu ◽  
Qiuxia Lin ◽  
Binhui Xie ◽  
Zhengming Ding ◽  
...  
Keyword(s):  
Domain Adaptation ◽  
Feature Learning ◽  
Feature Representations ◽  
Domain Specific ◽  
Cross Domain ◽  
Learning Tasks ◽  
Domain Learning ◽  
High Level ◽  
High Level Feature ◽  
Target Data

Tremendous research efforts have been made to thrive deep domain adaptation (DA) by seeking domain-invariant features. Most existing deep DA models only focus on aligning feature representations of task-specific layers across domains while integrating a totally shared convolutional architecture for source and target. However, we argue that such strongly-shared convolutional layers might be harmful for domain-specific feature learning when source and target data distribution differs to a large extent. In this paper, we relax a shared-convnets assumption made by previous DA methods and propose a Domain Conditioned Adaptation Network (DCAN), which aims to excite distinct convolutional channels with a domain conditioned channel attention mechanism. As a result, the critical low-level domain-dependent knowledge could be explored appropriately. As far as we know, this is the first work to explore the domain-wise convolutional channel activation for deep DA networks. Moreover, to effectively align high-level feature distributions across two domains, we further deploy domain conditioned feature correction blocks after task-specific layers, which will explicitly correct the domain discrepancy. Extensive experiments on three cross-domain benchmarks demonstrate the proposed approach outperforms existing methods by a large margin, especially on very tough cross-domain learning tasks.

scholarly journals Multimodal Linear Discriminant Analysis via Structural Sparsity

2017 ◽  
Author(s):  
Yu Zhang ◽  
Yuan Jiang
Keyword(s):  
Discriminant Analysis ◽  
Linear Discriminant Analysis ◽  
Real World ◽  
Specific Class ◽  
Linear Discriminant ◽  
Class Separability ◽  
Dimensionality Reduction Technique ◽  
Real World Applications ◽  
Data Points ◽  
Real World Datasets

Linear discriminant analysis (LDA) is a widely used supervised dimensionality reduction technique. Even though the LDA method has many real-world applications, it has some limitations such as the single-modal problem that each class follows a normal distribution. To solve this problem, we propose a method called multimodal linear discriminant analysis (MLDA). By generalizing the between-class and within-class scatter matrices, the MLDA model can allow each data point to have its own class mean which is called the instance-specific class mean. Then in each class, data points which share the same or similar instance-specific class means are considered to form one cluster or modal. In order to learn the instance-specific class means, we use the ratio of the proposed generalized between-class scatter measure over the proposed generalized within-class scatter measure, which encourages the class separability, as a criterion. The observation that each class will have a limited number of clusters inspires us to use a structural sparse regularizor to control the number of unique instance-specific class means in each class. Experiments on both synthetic and real-world datasets demonstrate the effectiveness of the proposed MLDA method.

scholarly journals Understanding Generalization in Neural Networks for Robustness against Adversarial Vulnerabilities

2020 ◽  
Vol 34 (10) ◽  
pp. 13714-13715
Author(s):  
Subhajit Chaudhury
Keyword(s):  
Neural Networks ◽  
Computer Vision ◽  
Real World ◽  
Speech Processing ◽  
State Of The Art ◽  
Feature Learning ◽  
Equivalent Problem ◽  
Tremendous Progress ◽  
Real World Applications ◽  
Out Of Sample

Neural networks have contributed to tremendous progress in the domains of computer vision, speech processing, and other real-world applications. However, recent studies have shown that these state-of-the-art models can be easily compromised by adding small imperceptible perturbations. My thesis summary frames the problem of adversarial robustness as an equivalent problem of learning suitable features that leads to good generalization in neural networks. This is motivated from learning in humans which is not trivially fooled by such perturbations due to robust feature learning which shows good out-of-sample generalization.

scholarly journals A repository of real-world datasets for data-driven evolutionary multiobjective optimization

2019 ◽  
Vol 6 (1) ◽  
pp. 189-197 ◽  
Author(s):  
Cheng He ◽  
Ye Tian ◽  
Handing Wang ◽  
Yaochu Jin
Keyword(s):  
Multiobjective Optimization ◽  
Real World ◽  
Optimization Problems ◽  
Data Driven ◽  
Benchmark Problems ◽  
Computational Costs ◽  
Evolutionary Multiobjective Optimization ◽  
Multiobjective Optimization Problems ◽  
Real World Applications ◽  
Real World Datasets

Abstract Many real-world optimization applications have more than one objective, which are modeled as multiobjective optimization problems. Generally, those complex objective functions are approximated by expensive simulations rather than cheap analytic functions, which have been formulated as data-driven multiobjective optimization problems. The high computational costs of those problems pose great challenges to existing evolutionary multiobjective optimization algorithms. Unfortunately, there have not been any benchmark problems reflecting those challenges yet. Therefore, we carefully select seven benchmark multiobjective optimization problems from real-world applications, aiming to promote the research on data-driven evolutionary multiobjective optimization by suggesting a set of benchmark problems extracted from various real-world optimization applications.

scholarly journals Selective oversampling approach for strongly imbalanced data

2021 ◽  
Vol 7 ◽  
pp. e604
Author(s):  
Peter Gnip ◽  
Liberios Vokorokos ◽  
Peter Drotár
Keyword(s):  
Outlier Detection ◽  
Real World ◽  
State Of The Art ◽  
Imbalanced Data ◽  
Prediction Performance ◽  
Classifier Performance ◽  
Real World Applications ◽  
Real World Datasets ◽  
Synthetic Datasets ◽  
Representative Samples

Challenges posed by imbalanced data are encountered in many real-world applications. One of the possible approaches to improve the classifier performance on imbalanced data is oversampling. In this paper, we propose the new selective oversampling approach (SOA) that first isolates the most representative samples from minority classes by using an outlier detection technique and then utilizes these samples for synthetic oversampling. We show that the proposed approach improves the performance of two state-of-the-art oversampling methods, namely, the synthetic minority oversampling technique and adaptive synthetic sampling. The prediction performance is evaluated on four synthetic datasets and four real-world datasets, and the proposed SOA methods always achieved the same or better performance than other considered existing oversampling methods.

scholarly journals Using Power-Law Degree Distribution to Accelerate PageRank

2012 ◽  
Vol 1 (2) ◽  
pp. 63-70
Author(s):  
Zhaoyan Jin ◽  
Quanyuan Wu
Keyword(s):  
World Wide Web ◽  
Power Law ◽  
Real World ◽  
Degree Distribution ◽  
World Wide ◽  
Power Law Distribution ◽  
The World ◽  
Real World Applications ◽  
Real World Datasets ◽  
Law Degree

The PageRank vector of a network is very important, for it can reflect the importance of a Web page in the World Wide Web, or of a people in a social network. However, with the growth of the World Wide Web and social networks, it needs more and more time to compute the PageRank vector of a network. In many real-world applications, the degree and PageRank distributions of these complex networks conform to the Power-Law distribution. This paper utilizes the degree distribution of a network to initialize its PageRank vector, and presents a Power-Law degree distribution accelerating algorithm of PageRank computation. Experiments on four real-world datasets show that the proposed algorithm converges more quickly than the original PageRank algorithm.DOI: 10.18495/comengapp.12.063070

scholarly journals A Noval Weighted Meta Graph Method for Classification in Heterogeneous Information Networks

2020 ◽  
Vol 10 (5) ◽  
pp. 1603
Author(s):  
Jinli Zhang ◽  
Tong Li ◽  
Zongli Jiang ◽  
Xiaohua Hu ◽  
Ali Jazayeri
Keyword(s):  
Real World ◽  
Structural Features ◽  
Information Networks ◽  
Heterogeneous Information ◽  
Heterogeneous Information Networks ◽  
Real World Applications ◽  
Different Types ◽  
Multiple Challenges ◽  
Real World Datasets

There has been increasing interest in the analysis and mining of Heterogeneous Information Networks (HINs) and the classification of their components in recent years. However, there are multiple challenges associated with distinguishing different types of objects in HINs in real-world applications. In this paper, a novel framework is proposed for the weighted Meta graph-based Classification of Heterogeneous Information Networks (MCHIN) to address these challenges. The proposed framework has several appealing properties. In contrast to other proposed approaches, MCHIN can fully compute the weights of different meta graphs and mine the latent structural features of different nodes by using these weighted meta graphs. Moreover, MCHIN significantly enlarges the training sets by introducing the concept of Extension Meta Graphs in HINs. The extension meta graphs are used to augment the semantic relationship among the source objects. Finally, based on the ranking distribution of objects, MCHIN groups the objects into pre-specified classes. We verify the performance of MCHIN on three real-world datasets. As is shown and discussed in the results section, the proposed framework can effectively outperform the baselines algorithms.

Supervised High-Level Feature Learning With Label Consistencies for Object Recognition

2020 ◽  
Vol 58 (7) ◽  
pp. 4501-4516 ◽  
Author(s):  
Yuebin Wang ◽  
Xun Zhou ◽  
Honglei Yang ◽  
Liqiang Zhang ◽  
Suhong Liu ◽  
...  
Keyword(s):  
Object Recognition ◽  
Feature Learning ◽  
High Level ◽  
High Level Feature

scholarly journals Robust Self-Weighted Multi-View Projection Clustering

2020 ◽  
Vol 34 (04) ◽  
pp. 6110-6117
Author(s):  
Beilei Wang ◽  
Yun Xiao ◽  
Zhihui Li ◽  
Xuanhong Wang ◽  
Xiaojiang Chen ◽  
...  
Keyword(s):  
Real World ◽  
State Of The Art ◽  
High Data ◽  
Real World Applications ◽  
Noise Clustering ◽  
Real World Datasets ◽  
Synthetic Datasets ◽  
Suppress Noise ◽  
Optimal Graph ◽  
Structure Graph

Many real-world applications involve data collected from different views and with high data dimensionality. Furthermore, multi-view data always has unavoidable noise. Clustering on this kind of high-dimensional and noisy multi-view data remains a challenge due to the curse of dimensionality and ineffective de-noising and integration of multiple views. Aiming at this problem, in this paper, we propose a Robust Self-weighted Multi-view Projection Clustering (RSwMPC) based on ℓ2,1-norm, which can simultaneously reduce dimensionality, suppress noise and learn local structure graph. Then the obtained optimal graph can be directly used for clustering while no further processing is required. In addition, a new method is introduced to automatically learn the optimal weight of each view with no need to generate additional parameters to adjust the weight. Extensive experimental results on different synthetic datasets and real-world datasets demonstrate that the proposed algorithm outperforms other state-of-the-art methods on clustering performance and robustness.

scholarly journals Systems AI: A Declarative Learning Based Programming Perspective

2018 ◽  
Author(s):  
Parisa Kordjamshidi ◽  
Dan Roth ◽  
Kristian Kersting
Keyword(s):  
Machine Learning ◽  
Real World ◽  
Data Driven ◽  
Real World Data ◽  
Future Directions ◽  
Knowledge Representations ◽  
Real World Applications ◽  
Need To Evaluate ◽  
Declarative Learning ◽  
High Level

Data-driven approaches are becoming dominant problem-solving techniques in many areas of research and industry. Unfortunately, current technologies do not make such techniques easy to use for application experts who are not fluent in machine learning nor for machine learning experts who aim at testing ideas on real-world data and need to evaluate those as a part of an end-to-end system. We review key efforts made by various AI communities to provide languages for high-level abstractions over learning and reasoning techniques needed for designing complex AI systems. We classify the existing frameworks based on the type of techniques as well as the data and knowledge representations they use, provide a comparative study of the way they address the challenges of programming real-world applications, and highlight some shortcomings and future directions.

Sign in / Sign up

Export Citation Format

Share Document