arXiv subjects

Hao Xue

Publications and source records attributed to Hao Xue.

107 records · Page 6Linked to original sources

arXiv2022

Leveraging Language Foundation Models for Human Mobility Forecasting

In this paper, we propose a novel pipeline that leverages language foundation models for temporal sequential pattern mining, such as for human mobility forecasting tasks. For example, in the task of predicting Place-of-Interest (POI) customer flows, typically the number of visits is extracted from historical logs, and only the numerical data are used to predict visitor flows. In this research, we perform the forecasting task directly on the natural language input that includes all kinds of information such as numerical values and contextual semantic information. Specific prompts are introduced to transform numerical temporal sequences into sentences so that existing language models can be directly applied. We design an AuxMobLCast pipeline for predicting the number of visitors in each POI, integrating an auxiliary POI category classification task with the encoder-decoder architecture. This research provides empirical evidence of the effectiveness of the proposed AuxMobLCast pipeline to discover sequential patterns in mobility forecasting tasks. The results, evaluated on three real-world datasets, demonstrate that pre-trained language foundation models also have good performance in forecasting temporal sequences. This study could provide visionary insights and lead to new research directions for predicting human mobility.

cs.LG↗

arXiv2022

COCOA: Cross Modality Contrastive Learning for Sensor Data

Self-Supervised Learning (SSL) is a new paradigm for learning discriminative representations without labelled data and has reached comparable or even state-of-the-art results in comparison to supervised counterparts. Contrastive Learning (CL) is one of the most well-known approaches in SSL that attempts to learn general, informative representations of data. CL methods have been mostly developed for applications in computer vision and natural language processing where only a single sensor modality is used. A majority of pervasive computing applications, however, exploit data from a range of different sensor modalities. While existing CL methods are limited to learning from one or two data sources, we propose COCOA (Cross mOdality COntrastive leArning), a self-supervised model that employs a novel objective function to learn quality representations from multisensor data by computing the cross-correlation between different data modalities and minimizing the similarity between irrelevant instances. We evaluate the effectiveness of COCOA against eight recently introduced state-of-the-art self-supervised models, and two supervised baselines across five public datasets. We show that COCOA achieves superior classification performance to all other approaches. Also, COCOA is far more label-efficient than the other baselines including the fully supervised model using only one-tenth of available labelled data.

cs.CV↗

arXiv2022

Beyond Just Vision: A Review on Self-Supervised Representation Learning on Multimodal and Temporal Data

Recently, Self-Supervised Representation Learning (SSRL) has attracted much attention in the field of computer vision, speech, natural language processing (NLP), and recently, with other types of modalities, including time series from sensors. The popularity of self-supervised learning is driven by the fact that traditional models typically require a huge amount of well-annotated data for training. Acquiring annotated data can be a difficult and costly process. Self-supervised methods have been introduced to improve the efficiency of training data through discriminative pre-training of models using supervisory signals that have been freely obtained from the raw data. Unlike existing reviews of SSRL that have pre-dominately focused upon methods in the fields of CV or NLP for a single modality, we aim to provide the first comprehensive review of multimodal self-supervised learning methods for temporal data. To this end, we 1) provide a comprehensive categorization of existing SSRL methods, 2) introduce a generic pipeline by defining the key components of a SSRL framework, 3) compare existing models in terms of their objective function, network architecture and potential applications, and 4) review existing multimodal techniques in each category and various modalities. Finally, we present existing weaknesses and future opportunities. We believe our work develops a perspective on the requirements of SSRL in domains that utilise multimodal and/or temporal data

cs.LG↗

arXiv2021

Event-Aware Multimodal Mobility Nowcasting

As a decisive part in the success of Mobility-as-a-Service (MaaS), spatio-temporal predictive modeling for crowd movements is a challenging task particularly considering scenarios where societal events drive mobility behavior deviated from the normality. While tremendous progress has been made to model high-level spatio-temporal regularities with deep learning, most, if not all of the existing methods are neither aware of the dynamic interactions among multiple transport modes nor adaptive to unprecedented volatility brought by potential societal events. In this paper, we are therefore motivated to improve the canonical spatio-temporal network (ST-Net) from two perspectives: (1) design a heterogeneous mobility information network (HMIN) to explicitly represent intermodality in multimodal mobility; (2) propose a memory-augmented dynamic filter generator (MDFG) to generate sequence-specific parameters in an on-the-fly fashion for various scenarios. The enhanced event-aware spatio-temporal network, namely EAST-Net, is evaluated on several real-world datasets with a wide variety and coverage of societal events. Both quantitative and qualitative experimental results verify the superiority of our approach compared with the state-of-the-art baselines. Code and data are published on https://github.com/underdoc-wang/EAST-Net.

cs.LG↗

arXiv2021

Translating Human Mobility Forecasting through Natural Language Generation

Existing human mobility forecasting models follow the standard design of the time-series prediction model which takes a series of numerical values as input to generate a numerical value as a prediction. Although treating this as a regression problem seems straightforward, incorporating various contextual information such as the semantic category information of each Place-of-Interest (POI) is a necessary step, and often the bottleneck, in designing an effective mobility prediction model. As opposed to the typical approach, we treat forecasting as a translation problem and propose a novel forecasting through a language generation pipeline. The paper aims to address the human mobility forecasting problem as a language translation task in a sequence-to-sequence manner. A mobility-to-language template is first introduced to describe the numerical mobility data as natural language sentences. The core intuition of the human mobility forecasting translation task is to convert the input mobility description sentences into a future mobility description from which the prediction target can be obtained. Under this pipeline, a two-branch network, SHIFT (Translating Human Mobility Forecasting), is designed. Specifically, it consists of one main branch for language generation and one auxiliary branch to directly learn mobility patterns. During the training, we develop a momentum mode for better connecting and training the two branches. Extensive experiments on three real-world datasets demonstrate that the proposed SHIFT is effective and presents a new revolutionary approach to forecasting human mobility.

cs.CL↗

arXiv2021

PIETS: Parallelised Irregularity Encoders for Forecasting with Heterogeneous Time-Series

Heterogeneity and irregularity of multi-source data sets present a significant challenge to time-series analysis. In the literature, the fusion of multi-source time-series has been achieved either by using ensemble learning models which ignore temporal patterns and correlation within features or by defining a fixed-size window to select specific parts of the data sets. On the other hand, many studies have shown major improvement to handle the irregularity of time-series, yet none of these studies has been applied to multi-source data. In this work, we design a novel architecture, PIETS, to model heterogeneous time-series. PIETS has the following characteristics: (1) irregularity encoders for multi-source samples that can leverage all available information and accelerate the convergence of the model; (2) parallelised neural networks to enable flexibility and avoid information overwhelming; and (3) attention mechanism that highlights different information and gives high importance to the most related data. Through extensive experiments on real-world data sets related to COVID-19, we show that the proposed architecture is able to effectively model heterogeneous temporal data and outperforms other state-of-the-art approaches in the prediction task.

cs.LG↗

arXiv2021

MobTCast: Leveraging Auxiliary Trajectory Forecasting for Human Mobility Prediction

Human mobility prediction is a core functionality in many location-based services and applications. However, due to the sparsity of mobility data, it is not an easy task to predict future POIs (place-of-interests) that are going to be visited. In this paper, we propose MobTCast, a Transformer-based context-aware network for mobility prediction. Specifically, we explore the influence of four types of context in the mobility prediction: temporal, semantic, social and geographical contexts. We first design a base mobility feature extractor using the Transformer architecture, which takes both the history POI sequence and the semantic information as input. It handles both the temporal and semantic contexts. Based on the base extractor and the social connections of a user, we employ a self-attention module to model the influence of the social context. Furthermore, unlike existing methods, we introduce a location prediction branch in MobTCast as an auxiliary task to model the geographical context and predict the next location. Intuitively, the geographical distance between the location of the predicted POI and the predicted location from the auxiliary branch should be as close as possible. To reflect this relation, we design a consistency loss to further improve the POI prediction performance. In our experimental results, MobTCast outperforms other state-of-the-art next POI prediction methods. Our approach illustrates the value of including different types of context in next POI prediction.

cs.LG↗

arXiv2021

Generative Adversarial Networks for Spatio-temporal Data: A Survey

Generative Adversarial Networks (GANs) have shown remarkable success in producing realistic-looking images in the computer vision area. Recently, GAN-based techniques are shown to be promising for spatio-temporal-based applications such as trajectory prediction, events generation and time-series data imputation. While several reviews for GANs in computer vision have been presented, no one has considered addressing the practical applications and challenges relevant to spatio-temporal data. In this paper, we have conducted a comprehensive review of the recent developments of GANs for spatio-temporal data. We summarise the application of popular GAN architectures for spatio-temporal data and the common practices for evaluating the performance of spatio-temporal applications with GANs. Finally, we point out future research directions to benefit researchers in this area.

cs.LG↗

arXiv2021

Exploring Self-Supervised Representation Ensembles for COVID-19 Cough Classification

The usage of smartphone-collected respiratory sound, trained with deep learning models, for detecting and classifying COVID-19 becomes popular recently. It removes the need for in-person testing procedures especially for rural regions where related medical supplies, experienced workers, and equipment are limited. However, existing sound-based diagnostic approaches are trained in a fully supervised manner, which requires large scale well-labelled data. It is critical to discover new methods to leverage unlabelled respiratory data, which can be obtained more easily. In this paper, we propose a novel self-supervised learning enabled framework for COVID-19 cough classification. A contrastive pre-training phase is introduced to train a Transformer-based feature encoder with unlabelled data. Specifically, we design a random masking mechanism to learn robust representations of respiratory sounds. The pre-trained feature encoder is then fine-tuned in the downstream phase to perform cough classification. In addition, different ensembles with varied random masking rates are also explored in the downstream phase. Through extensive evaluations, we demonstrate that the proposed contrastive pre-training, the random masking mechanism, and the ensemble architecture contribute to improving cough classification performance.

cs.SD↗

arXiv2021

TERMCast: Temporal Relation Modeling for Effective Urban Flow Forecasting

Urban flow forecasting is a challenging task, given the inherent periodic characteristics of urban flow patterns. To capture the periodicity, existing urban flow prediction approaches are often designed with closeness, period, and trend components extracted from the urban flow sequence. However, these three components are often considered separately in the prediction model. These components have not been fully explored together and simultaneously incorporated in urban flow forecasting models. We introduce a novel urban flow forecasting architecture, TERMCast. A Transformer based long-term relation prediction module is explicitly designed to discover the periodicity and enable the three components to be jointly modeled This module predicts the periodic relation which is then used to yield the predicted urban flow tensor. To measure the consistency of the predicted periodic relation vector and the relation vector inferred from the predicted urban flow tensor, we propose a consistency module. A consistency loss is introduced in the training process to further improve the prediction performance. Through extensive experiments on three real-world datasets, we demonstrate that TERMCast outperforms multiple state-of-the-art methods. The effectiveness of each module in TERMCast has also been investigated.

cs.AI↗

arXiv2021

Time Series Change Point Detection with Self-Supervised Contrastive Predictive Coding

Change Point Detection (CPD) methods identify the times associated with changes in the trends and properties of time series data in order to describe the underlying behaviour of the system. For instance, detecting the changes and anomalies associated with web service usage, application usage or human behaviour can provide valuable insights for downstream modelling tasks. We propose a novel approach for self-supervised Time Series Change Point detection method based onContrastivePredictive coding (TS-CP^2). TS-CP^2 is the first approach to employ a contrastive learning strategy for CPD by learning an embedded representation that separates pairs of embeddings of time adjacent intervals from pairs of interval embeddings separated across time. Through extensive experiments on three diverse, widely used time series datasets, we demonstrate that our method outperforms five state-of-the-art CPD methods, which include unsupervised and semi-supervisedapproaches. TS-CP^2 is shown to improve the performance of methods that use either handcrafted statistical or temporal features by 79.4% and deep learning-based methods by 17.0% with respect to the F1-score averaged across the three datasets.

cs.LG↗

arXiv2020

Scene Gated Social Graph: Pedestrian Trajectory Prediction Based on Dynamic Social Graphs and Scene Constraints

Pedestrian trajectory prediction is valuable for understanding human motion behaviors and it is challenging because of the social influence from other pedestrians, the scene constraints and the multimodal possibilities of predicted trajectories. Most existing methods only focus on two of the above three key elements. In order to jointly consider all these elements, we propose a novel trajectory prediction method named Scene Gated Social Graph (SGSG). In the proposed SGSG, dynamic graphs are used to describe the social relationship among pedestrians. The social and scene influences are taken into account through the scene gated social graph features which combine the encoded social graph features and semantic scene features. In addition, a VAE module is incorporated to learn the scene gated social feature and sample latent variables for generating multiple trajectories that are socially and environmentally acceptable. We compare our SGSG against twenty state-of-the-art pedestrian trajectory prediction methods and the results show that the proposed method achieves superior performance on two widely used trajectory prediction benchmarks.

cs.CV↗

arXiv2020

Critical scaling for an anisotropic percolation system on $\mathbb{Z}^2$

In this article, we consider an anisotropic finite-range bond percolation model on $\mathbb{Z}^2$. On each horizontal layer $\{(x,i): x \in \mathbb{Z}\}$ we have edges $\langle(x, i),(y, i)\rangle$ for $1 \leq |x - y| \leq N$. There are also vertical edges connecting two nearest neighbor vertices on distinct layers $\langle(x, i), (x, i+1)\rangle$ for $x, i \in\mathbb{Z}$. On this graph we consider the following anisotropic independent percolation model: horizontal edges are open with probability $1/(2N)$, while vertical edges are open with probability $ε$ to be suitably tuned as $N$ grows to infinity. The main result tells that if $ε=κN^{-2/5}$, we see a phase transition in $κ$: positive and finite constants $C_1, C_2$ exist so that there is no percolation if $κ< C_1$ while percolation occurs for $κ> C_2$. The question is motivated by a result on the analogous layered ferromagnetic Ising model at mean field critical temperature [J. Stat. Phys. (2015), 161, 91-123] where the authors showed the existence of multiple Gibbs measures for a fixed value of the vertical interaction and conjectured a change of behavior in $κ$ when the vertical interaction suitably vanishes as $κγ^b$, where $1/γ$ is the range of the horizontal interaction. For the product percolation model we have a value of $b$ that differs from what was conjectured in that paper. The proof relies on the analysis of the scaling limit of the critical branching random walk that dominates the growth process restricted to each horizontal layer and a careful analysis of the true horizontal growth process. This is inspired by works on the long range contact process [Probab. Th. Rel. Fields (1995), 102, 519-545]. A renormalization scheme is used for the percolative regime.

math.PR↗

arXiv2020

Take a NAP: Non-Autoregressive Prediction for Pedestrian Trajectories

Pedestrian trajectory prediction is a challenging task as there are three properties of human movement behaviors which need to be addressed, namely, the social influence from other pedestrians, the scene constraints, and the multimodal (multiroute) nature of predictions. Although existing methods have explored these key properties, the prediction process of these methods is autoregressive. This means they can only predict future locations sequentially. In this paper, we present NAP, a non-autoregressive method for trajectory prediction. Our method comprises specifically designed feature encoders and a latent variable generator to handle the three properties above. It also has a time-agnostic context generator and a time-specific context generator for non-autoregressive prediction. Through extensive experiments that compare NAP against several recent methods, we show that NAP has state-of-the-art trajectory prediction performance.

cs.CV↗

arXiv2019

Lateral Heterojunction BaTiO3/AlGaN Diodes with >8MV/cm Breakdown Field

In this paper, we report enhanced breakdown characteristics of Pt/BaTiO3/Al0.58Ga0.42N lateral heterojunction diodes compared to Pt/Al0.58Ga0.42N Schottky diodes. BaTiO3, an extreme dielectric constant material, has been used, in this study, as dielectric material under the anode to significantly reduce the peak electric field at the anode edge of the heterojunction diode such that the observed average breakdown field was higher than 8 MV/cm, achieved for devices with anode to cathode spacing less than 0.2 microns. Control Schottky anode devices (Pt/Al0.58Ga0.42N) fabricated on the same sample displayed an average breakdown field around 4 MV/cm for devices with similar dimensions. While both breakdown fields are significantly higher than those exhibited by incumbent technologies such as GaN-based devices, BaTiO3 can enable more effective utilization of the higher breakdown fields available in ultra-wide bandgap materials by proper electric field management. This demonstration thus lays the groundwork needed to realize ultra-scaled lateral devices with significantly improved breakdown characteristics.

physics.app-ph↗

arXiv2019

Compositionally graded contact layers for MOCVD grown high Al-content AlGaN transistors

In this letter, we design and demonstrate an improved MOCVD grown reverse Al-composition graded contact layer to achieve low resistance contact to MOCVD grown ultra-wide bandgap (UWBG) Al0.70Ga0.30N channel metal semiconductor field-effect transistors (MESFETs). Increasing the thickness of the reverse graded layer was found to improve contact layer resistance significantly, leading to contact resistance of 3.3x10^-5 Ohm.cm2. Devices with gate length, LG, of 0.6 microns and source-drain spacing, LSD, of 1.5 microns displayed a maximum current density, IDSMAX, of 635 mA/mm with an applied gate voltage, VGS, of +2 V. Breakdown measurements on transistors with gate to drain spacing, LGD, of 770 nm had breakdown voltage greater than 220 , corresponding to minimum breakdown field of 2.86 MV/cm. This work provides a framework for the design of low resistance contacts to MOCVD grown high Al-content AlxGa1-xN channel transistors.

physics.app-ph↗

arXiv2018

Liquid Epitaxial Growth of Two-dimensional Non-layer structured hybrid Perovskite

The hybrid perovskite CH3NH3PbX3 (X= Cl, Br, I) is a promising material for developing novel optoelectronic devices. Owing to the intrinsic non-layer structure, it remains challenging to synthesize molecularly thin CH3NH3PbX3 with large size. Here, we report a low-cost and highly efficient fabrication route to obtain large-scale single-crystalline 2D CH3NH3PbX3 perovskites on a mica substrate via liquid epitaxy. The 2D perovskite is characterized as 8 nm in thickness and hundreds of micrometers in lateral size. First-principles calculations suggest the strong potassium-halogen interactions at the perovskite/mica interface lower the interface energy of perovskites, driving their fast in-plane growth. Spectroscopic investigations reveal 2D CH3NH3PbBr3 possess small exciton binding energy of 30 meV, allowing a superior visible-light photodetector with a photoresponsivity of 126 A/W and a bandwidth exceeded 80 kHz. These features demonstrate that liquid epitaxy is a bottom-up approach to fabricate the non-layer structured 2D perovskites, which offer a new material platform for the device applications and fundamental investigations.

cond-mat.mtrl-sci↗