MDTP

Traffic prediction has drawn increasing attention for its ubiquitous real-life applications in traffic management, urban computing, public safety, and so on. Recently, the availability of massive trajectory data and the success of deep learning motivate a plethora of deep traffic prediction studies. However, the existing neural-network-based approaches tend to ignore the correlations between multiple types of moving objects located in the same spatio-temporal traffic area, which is suboptimal for traffic prediction analytics. In this paper, we propose a multi-source deep traffic prediction framework over spatio-temporal trajectory data, termed as MDTP. The framework includes two phases: spatio-temporal feature modeling and multi-source bridging. We present an enhanced graph convolutional network (GCN) model combined with long short-term memory network (LSTM) to capture the spatial dependencies and temporal dynamics of traffic in the feature modeling phase. In the multi-source bridging phase, we propose two methods, Sum and Concat, to connect the learned features from different trajectory data sources. Extensive experiments on two real-life datasets show that MDTP i) has superior efficiency, compared with classical time-series methods, machine learning methods, and state-of-the-art neural-network-based approaches; ii) offers a significant performance improvement over the single-source traffic prediction approach; and iii) performs traffic predictions in seconds even on tens of millions of trajectory data. we develop MDTP + , a user-friendly interactive system to demonstrate traffic prediction analysis.

Download Full-text

FetNet: a recurrent convolutional network for occlusion identification in fetoscopic videos

International Journal of Computer Assisted Radiology and Surgery ◽

10.1007/s11548-020-02169-0 ◽

2020 ◽

Vol 15 (5) ◽

pp. 791-801 ◽

Cited By ~ 1

Author(s):

Sophia Bano ◽

Francisco Vasconcelos ◽

Emmanuel Vander Poorten ◽

Tom Vercauteren ◽

Sebastien Ourselin ◽

...

Keyword(s):

Neural Network ◽

Network Architecture ◽

Laser Photocoagulation ◽

Short Term Memory ◽

Total Duration ◽

Frame Rate ◽

Superior Performance ◽

Computer Assisted ◽

Convolutional Network ◽

Spatio Temporal

Abstract Purpose Fetoscopic laser photocoagulation is a minimally invasive surgery for the treatment of twin-to-twin transfusion syndrome (TTTS). By using a lens/fibre-optic scope, inserted into the amniotic cavity, the abnormal placental vascular anastomoses are identified and ablated to regulate blood flow to both fetuses. Limited field-of-view, occlusions due to fetus presence and low visibility make it difficult to identify all vascular anastomoses. Automatic computer-assisted techniques may provide better understanding of the anatomical structure during surgery for risk-free laser photocoagulation and may facilitate in improving mosaics from fetoscopic videos. Methods We propose FetNet, a combined convolutional neural network (CNN) and long short-term memory (LSTM) recurrent neural network architecture for the spatio-temporal identification of fetoscopic events. We adapt an existing CNN architecture for spatial feature extraction and integrated it with the LSTM network for end-to-end spatio-temporal inference. We introduce differential learning rates during the model training to effectively utilising the pre-trained CNN weights. This may support computer-assisted interventions (CAI) during fetoscopic laser photocoagulation. Results We perform quantitative evaluation of our method using 7 in vivo fetoscopic videos captured from different human TTTS cases. The total duration of these videos was 5551 s (138,780 frames). To test the robustness of the proposed approach, we perform 7-fold cross-validation where each video is treated as a hold-out or test set and training is performed using the remaining videos. Conclusion FetNet achieved superior performance compared to the existing CNN-based methods and provided improved inference because of the spatio-temporal information modelling. Online testing of FetNet, using a Tesla V100-DGXS-32GB GPU, achieved a frame rate of 114 fps. These results show that our method could potentially provide a real-time solution for CAI and automating occlusion and photocoagulation identification during fetoscopic procedures.

Download Full-text

SAST-GNN: A Self-Attention Based Spatio-Temporal Graph Neural Network for Traffic Prediction

Database Systems for Advanced Applications - Lecture Notes in Computer Science ◽

10.1007/978-3-030-59410-7_49 ◽

2020 ◽

pp. 707-714

Author(s):

Yi Xie ◽

Yun Xiong ◽

Yangyong Zhu

Keyword(s):

Neural Network ◽

Traffic Prediction ◽

Temporal Graph ◽

Spatio Temporal

Download Full-text

Identifying Foreign Tourists’ Nationality from Mobility Traces via LSTM Neural Network and Location Embeddings

Applied Sciences ◽

10.3390/app9142861 ◽

2019 ◽

Vol 9 (14) ◽

pp. 2861 ◽

Cited By ~ 2

Author(s):

Alessandro Crivellari ◽

Euro Beinat

Keyword(s):

Neural Network ◽

Motion Tracking ◽

Large Scale ◽

Short Term Memory ◽

Human Mobility ◽

User Profiling ◽

Personalized Recommendation ◽

Mobility Patterns ◽

Trajectory Data ◽

Short Term

The interest in human mobility analysis has increased with the rapid growth of positioning technology and motion tracking, leading to a variety of studies based on trajectory recordings. Mapping the routes that people commonly perform was revealed to be very useful for location-based service applications, where individual mobility behaviors can potentially disclose meaningful information about each customer and be fruitfully used for personalized recommendation systems. This paper tackles a novel trajectory labeling problem related to the context of user profiling in “smart” tourism, inferring the nationality of individual users on the basis of their motion trajectories. In particular, we use large-scale motion traces of short-term foreign visitors as a way of detecting the nationality of individuals. This task is not trivial, relying on the hypothesis that foreign tourists of different nationalities may not only visit different locations, but also move in a different way between the same locations. The problem is defined as a multinomial classification with a few tens of classes (nationalities) and sparse location-based trajectory data. We hereby propose a machine learning-based methodology, consisting of a long short-term memory (LSTM) neural network trained on vector representations of locations, in order to capture the underlying semantics of user mobility patterns. Experiments conducted on a real-world big dataset demonstrate that our method achieves considerably higher performances than baseline and traditional approaches.

Download Full-text

Data-Driven Predictive Modeling of Neuronal Dynamics Using Long Short-Term Memory

Algorithms ◽

10.3390/a12100203 ◽

2019 ◽

Vol 12 (10) ◽

pp. 203

Author(s):

Benjamin Plaster ◽

Gautam Kumar

Keyword(s):

Neural Network ◽

Time Horizon ◽

Short Term Memory ◽

Temporal Dynamics ◽

Data Driven ◽

Brain Dynamics ◽

Short Term ◽

Term Memory ◽

Brain Functions ◽

Long Short Term Memory

Modeling brain dynamics to better understand and control complex behaviors underlying various cognitive brain functions have been of interest to engineers, mathematicians and physicists over the last several decades. With the motivation of developing computationally efficient models of brain dynamics to use in designing control-theoretic neurostimulation strategies, we have developed a novel data-driven approach in a long short-term memory (LSTM) neural network architecture to predict the temporal dynamics of complex systems over an extended long time-horizon in future. In contrast to recent LSTM-based dynamical modeling approaches that make use of multi-layer perceptrons or linear combination layers as output layers, our architecture uses a single fully connected output layer and reversed-order sequence-to-sequence mapping to improve short time-horizon prediction accuracy and to make multi-timestep predictions of dynamical behaviors. We demonstrate the efficacy of our approach in reconstructing the regular spiking to bursting dynamics exhibited by an experimentally-validated 9-dimensional Hodgkin-Huxley model of hippocampal CA1 pyramidal neurons. Through simulations, we show that our LSTM neural network can predict the multi-time scale temporal dynamics underlying various spiking patterns with reasonable accuracy. Moreover, our results show that the predictions improve with increasing predictive time-horizon in the multi-timestep deep LSTM neural network.

Download Full-text

ON MEMORY AND STRUCTURAL DYNAMISM IN EXCITABLE CELLULAR AUTOMATA WITH DEFENSIVE INHIBITION

International Journal of Bifurcation and Chaos ◽

10.1142/s0218127408020458 ◽

2008 ◽

Vol 18 (02) ◽

pp. 527-539 ◽

Cited By ~ 6

Author(s):

RAMÓN ALONSO-SANZ ◽

ANDREW ADAMATZKY

Keyword(s):

Cellular Automata ◽

Short Term Memory ◽

Temporal Dynamics ◽

Short Term ◽

Resting Cell ◽

Fixed Topology ◽

Dynamical Topology ◽

Novel Complex ◽

Spatio Temporal ◽

Update Rules

Commonly studied cellular automata are memoryless and have fixed topology of connections between cells. However by allowing updates of links and short-term memory in cells we may potentially discover novel complex regimes of spatio-temporal dynamics. Moreover, by adding memory and dynamical topology to state update rules we somehow forge elementary but nontraditional models of neurons networks (aka neuron layers in frontal parts). In the present paper, we demonstrate how this can be done on a self-inhibitory excitable cellular automata. These automata imitate a phenomenon of inhibition caused by hight-strength stimulus: a resting cell excites if there are one or two excited neighbors, the cell remains resting otherwise. We modify the automaton by allowing cells to have few-steps memories, and create links between neighboring cells removed or generated depending on the states of the cells.

Download Full-text

APPLICATION OF DISTRIBUTED PARAMETER CONTROL MODEL IN WILDLIFE DAMAGE MANAGEMENT

Mathematical Models and Methods in Applied Sciences ◽

10.1142/s0218202592000259 ◽

1992 ◽

Vol 02 (04) ◽

pp. 423-439 ◽

Cited By ~ 28

Author(s):

SUZANNE M. LENHART ◽

MAHADEV G. BHAT

Keyword(s):

Temporal Dynamics ◽

Real Life ◽

Optimal Solution ◽

Distributed Parameter ◽

Parameter Control ◽

Wildlife Damage Management ◽

Distributed Parameter Control ◽

Time Problem ◽

Wildlife Damage ◽

Spatio Temporal

A bioeconomic model for optimal control of wildlife damage by migratory small mammal populations is developed under the framework of a nonlinear distributed parameter control problem. The model first simulates the spatio-temporal dynamics of dispersal population by parabolic diffusive Volterra-Lotka partial differential equation and then optimizes a criterion function of present value combined costs of wildlife damage and harvesting. The existence of a unique optimal solution for a finite time problem is proved. An iterative procedure for numerical solution of the Optimality System with parabolic equations of opposite orientations is developed. The theoretical model is applied to a real life problem using biological and economic data for beaver populations under certain simplistic assumptions.

Download Full-text

An Incremental Class-Learning Approach with Acoustic Novelty Detection for Acoustic Event Recognition

Sensors ◽

10.3390/s21196622 ◽

2021 ◽

Vol 21 (19) ◽

pp. 6622

Author(s):

Barış Bayram ◽

Gökhan İnce

Keyword(s):

Neural Network ◽

Acoustic Signal ◽

Large Scale ◽

Short Term Memory ◽

Novelty Detection ◽

Event Recognition ◽

Acoustic Event ◽

Audio Features ◽

Spatio Temporal ◽

Self Learning

Acoustic scene analysis (ASA) relies on the dynamic sensing and understanding of stationary and non-stationary sounds from various events, background noises and human actions with objects. However, the spatio-temporal nature of the sound signals may not be stationary, and novel events may exist that eventually deteriorate the performance of the analysis. In this study, a self-learning-based ASA for acoustic event recognition (AER) is presented to detect and incrementally learn novel acoustic events by tackling catastrophic forgetting. The proposed ASA framework comprises six elements: (1) raw acoustic signal pre-processing, (2) low-level and deep audio feature extraction, (3) acoustic novelty detection (AND), (4) acoustic signal augmentations, (5) incremental class-learning (ICL) (of the audio features of the novel events) and (6) AER. The self-learning on different types of audio features extracted from the acoustic signals of various events occurs without human supervision. For the extraction of deep audio representations, in addition to visual geometry group (VGG) and residual neural network (ResNet), time-delay neural network (TDNN) and TDNN based long short-term memory (TDNN–LSTM) networks are pre-trained using a large-scale audio dataset, Google AudioSet. The performances of ICL with AND using Mel-spectrograms, and deep features with TDNNs, VGG, and ResNet from the Mel-spectrograms are validated on benchmark audio datasets such as ESC-10, ESC-50, UrbanSound8K (US8K), and an audio dataset collected by the authors in a real domestic environment.

Download Full-text

Data-Driven Predictive Modeling of Neuronal Dynamics using Long Short-Term Memory

10.20944/preprints201908.0155.v2 ◽

2019 ◽

Author(s):

Benjamin Plaster ◽

Gautam Kumar

Keyword(s):

Neural Network ◽

Time Horizon ◽

Short Term Memory ◽

Temporal Dynamics ◽

Data Driven ◽

Brain Dynamics ◽

Short Term ◽

Term Memory ◽

Brain Functions ◽

Long Short Term Memory

Modeling brain dynamics to better understand and control complex behaviors underlying various cognitive brain functions are of interests to engineers, mathematicians, and physicists from the last several decades. With a motivation of developing computationally efficient models of brain dynamics to use in designing control-theoretic neurostimulation strategies, we have developed a novel data-driven approach in a long short-term memory (LSTM) neural network architecture to predict the temporal dynamics of complex systems over an extended long time-horizon in future. In contrast to recent LSTM-based dynamical modeling approaches that make use of multi-layer perceptrons or linear combination layers as output layers, our architecture uses a single fully connected output layer and reversed-order sequence-to-sequence mapping to improve short time-horizon prediction accuracy and to make multi-timestep predictions of dynamical behaviors. We demonstrate the efficacy of our approach in reconstructing the regular spiking to bursting dynamics exhibited by an experimentally-validated 9-dimensional Hodgkin-Huxley model of hippocampal CA1 pyramidal neurons. Through simulations, we show that our LSTM neural network can predict the multi-time scale temporal dynamics underlying various spiking patterns with reasonable accuracy. Moreover, our results show that the predictions improve with increasing predictive time-horizon in the multi-timestep deep LSTM neural network.

Download Full-text