• Corpus ID: 52180364

Coupled IGMM-GANs for deep multimodal anomaly detection in human mobility data

@article{Gray2018CoupledIF,
  title={Coupled IGMM-GANs for deep multimodal anomaly detection in human mobility data},
  author={Kathryn Gray and Daniel Smolyak and Sarkhan Badirli and George O. Mohler},
  journal={ArXiv},
  year={2018},
  volume={abs/1809.02728}
}
Detecting anomalous activity in human mobility data has a number of applications including road hazard sensing, telematic based insurance, and fraud detection in taxi services and ride sharing. In this paper we address two challenges that arise in the study of anomalous human trajectories: 1) a lack of ground truth data on what defines an anomaly and 2) the dependence of existing methods on significant pre-processing and feature engineering. While generative adversarial networks seem like a… 

Figures and Tables from this paper

Collective Anomaly Detection for Multivariate Data using Generative Adversarial Networks

  • Chihiro MaruI. Kobayashi
  • Computer Science
    2020 International Conference on Computational Science and Computational Intelligence (CSCI)
  • 2020
TLDR
To correctly detect collective anomalies, an encoder is introduced into a GAN-based anomaly detection method to obtain the latent states of the real data and a sequence to sequence technique is adopted to both encoder and generator, recurrent neural network, and fully connected neural network for the discriminator.

Online Anomalous Trajectory Detection with Deep Generative Sequence Modeling

TLDR
This work proposes a novel model, namely Gaussian Mixture Variational Sequence AutoEncoder (GM-VSAE), that is able to capture complex sequential information enclosed in trajectories, discover different types of normal routes from trajectories and represent them in a continuous latent space, and support efficient online detection via trajectory generation.

A Novel and Efficient CVAE-GAN-Based Approach With Informative Manifold for Semi-Supervised Anomaly Detection

TLDR
A novel end-to-end deep network as an anomaly detector only trained on normal samples, which considers input categories, constructs, and exploits a useful manifold in an unsupervised manner and has a stronger reconstruction capability.

Brain Tumor Anomaly Detection via Latent Regularized Adversarial Network

TLDR
The semi-supervised anomaly detection model is proposed in which only healthy (normal) brain images are trained and achieves outperforms or comparable results to state-of-the-art supervised techniques.

Anomaly Detection with Adversarial Dual Autoencoders

TLDR
A GAN-based anomaly detection framework that consists of two autoencoders as generator and discriminator to increase training stability and employ discriminator reconstruction error as anomaly score for better detection performance is introduced.

DeepTEA: Effective and Efficient Online Time-dependent Trajectory Outlier Detection

TLDR
This paper proposes the deep-probabilistic-based time-dependent anomaly detection algorithm (DeepTEA), which employs deeplearning methods to obtain time- dependent outliners from a huge volume of trajectories, and can handle complex traffic conditions and detect outliners accurately.

DST-Predict: Predicting Individual Mobility Patterns From Mobile Phone GPS Data

TLDR
A novel multi-view sequence-to-sequence model that uses Convolutional Long-short term memory (ConvLSTM) where the past history of frequent visit patterns features is used to predict individuals’ future visit patterns in a multi-step manner is proposed.

Learning Disentangled Feature Representations For Anomaly Detection

  • Wei-Yu LeeY. Wang
  • Computer Science
    2020 IEEE International Conference on Image Processing (ICIP)
  • 2020
TLDR
A novel deep learning based method is proposed, which learns disentangled feature representations for separating semantic and visual appearance information, so that the anomaly of the input data can be determined based on its semantic features.

Quantifying the Tradeoff Between Cybersecurity and Location Privacy

TLDR
The experiment results suggest that applying Laplace mechanism to achieve high level of differential privacy in the context of location-based vehicle trips will result in low true-positive or high false-negative rate by the RNN, which diminishes the value of RNN as more anomalous trips will be classified as normal ones.

Where's Wally Now? Deep Generative and Discriminative Embeddings for Novelty Detection

TLDR
A generative novelty detection method is demonstrated, as one of the methods in the ND framework, whose performance exceeds that of all recent best-in-class generative ND methods.

References

SHOWING 1-10 OF 28 REFERENCES

GANomaly: Semi-Supervised Anomaly Detection via Adversarial Training

TLDR
This work introduces a novel anomaly detection model, by using a conditional generative adversarial network that jointly learns the generation of high-dimensional image space and the inference of latent space and shows the model efficacy and superiority over previous state-of-the-art approaches.

Efficient GAN-Based Anomaly Detection

TLDR
This work leverages recently developed GAN models for anomaly detection, and achieves state-of-the-art performance on image and network intrusion datasets, while being several hundred-fold faster at test time than the only published GAN-based method.

SenseGen: A deep learning architecture for synthetic sensor data generation

TLDR
A step towards generating sensory data that can pass a deep learning based discriminator model test is taken, and aDeep learning based architecture for synthesizing sensory data is presented, which comprises of a generator model, which is a stack of multiple Long-Short-Term-Memory networks and a Mixture Density Network (MDN).

Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery

TLDR
AnoGAN, a deep convolutional generative adversarial network is proposed to learn a manifold of normal anatomical variability, accompanying a novel anomaly scoring scheme based on the mapping from image space to a latent space.

Crowd sensing of traffic anomalies based on human mobility and social media

TLDR
This paper addresses the problem of detecting and describing traffic anomalies using crowd sensing with two forms of data, human mobility and social media, and identifies anomalies according to drivers' routing behavior on an urban road network.

Synthesizing realistic neural population activity patterns using Generative Adversarial Networks

TLDR
Spike-GAN provides a powerful, easy-to-use technique for generating realistic spiking neural activity and for describing the most relevant features of the large-scale neural population recordings studied in modern systems neuroscience.

Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

TLDR
A recurrent sequence-to-sequence model observes motion histories and predicts future behavior, using a novel pooling mechanism to aggregate information across people, and outperforms prior work in terms of accuracy, variety, collision avoidance, and computational complexity.

iBOAT: Isolation-Based Online Anomalous Trajectory Detection

TLDR
The proposed isolation-based online anomalous trajectory (iBOAT) is evaluated through extensive experiments on large-scale taxi data, and it shows that iBOAT achieves state-of-the-art performance, with a remarkable performance of the area under a curve (AUC) ≥ 0.99.

iBAT: detecting anomalous taxi trajectories from GPS traces

TLDR
An Isolation-Based Anomalous Trajectory (iBAT) detection method is proposed and the potential of iBAT in enabling innovative applications is demonstrated by using it for taxi driving fraud detection and road network change detection.

Generalization of Deep Neural Networks for Chest Pathology Classification in X-Rays Using Generative Adversarial Networks

TLDR
This work implements a generative adversarial network (GAN) to create artificial images based upon a modest sized labeled dataset to detect pathology across five classes of chest X-rays, and demonstrates that augmenting the original imbalanced dataset with GAN generated images improves performance of chest pathology classification.