arXiv subjects

Xiao Pan

Publications and source records attributed to Xiao Pan.

31 records · Page 2Linked to original sources

arXiv2022

Dynamic Gradient Reactivation for Backward Compatible Person Re-identification

We study the backward compatible problem for person re-identification (Re-ID), which aims to constrain the features of an updated new model to be comparable with the existing features from the old model in galleries. Most of the existing works adopt distillation-based methods, which focus on pushing new features to imitate the distribution of the old ones. However, the distillation-based methods are intrinsically sub-optimal since it forces the new feature space to imitate the inferior old feature space. To address this issue, we propose the Ranking-based Backward Compatible Learning (RBCL), which directly optimizes the ranking metric between new features and old features. Different from previous methods, RBCL only pushes the new features to find best-ranking positions in the old feature space instead of strictly alignment, and is in line with the ultimate goal of backward retrieval. However, the sharp sigmoid function used to make the ranking metric differentiable also incurs the gradient vanish issue, therefore stems the ranking refinement during the later period of training. To address this issue, we propose the Dynamic Gradient Reactivation (DGR), which can reactivate the suppressed gradients by adding dynamic computed constant during forward step. To further help targeting the best-ranking positions, we include the Neighbor Context Agents (NCAs) to approximate the entire old feature space during training. Unlike previous works which only test on the in-domain settings, we make the first attempt to introduce the cross-domain settings (including both supervised and unsupervised), which are more meaningful and difficult. The experimental results on all five settings show that the proposed RBCL outperforms previous state-of-the-art methods by large margins under all settings.

cs.CV↗

arXiv2022

Optimal Energy Management of Series Hybrid Electric Vehicles with Engine Start-Stop System

This paper develops energy management (EM) control for series hybrid electric vehicles (HEVs) that include an engine start-stop system (SSS). The objective of the control is to optimally split the energy between the sources of the powertrain and achieve fuel consumption minimization. In contrast to existing works, a fuel penalty is used to characterize more realistically SSS engine restarts, to enable more realistic design and testing of control algorithms. The paper first derives two important analytic results: a) analytic EM optimal solutions of fundamental and commonly used series HEV frameworks, and b) proof of optimality of charge sustaining operation in series HEVs. It then proposes a novel heuristic control strategy, the hysteresis power threshold strategy (HPTS), by amalgamating simple and effective control rules extracted from the suite of derived analytic EM optimal solutions. The decision parameters of the control strategy are small in number and freely tunable. The overall control performance can be fully optimized for different HEV parameters and driving cycles by a systematic tuning process, while also targeting charge sustaining operation. The performance of HPTS is evaluated and benchmarked against existing methodologies, including dynamic programming (DP) and a recently proposed state-of-the-art heuristic strategy. The results show the effectiveness and robustness of the HPTS and also indicate its potential to be used as the benchmark strategy for high fidelity HEV models, where DP is no longer applicable due to computational complexity.

eess.SY↗

arXiv2021

Systematic study of two-proton radioactivity half-lives based on a modified Gamow-like model

In the present work, we systematically study the two-proton (2p) radioactivity half-lives of nuclei close to the proton drip line within a modified Gamow-like model. Using this model, the calculated 2p radioactivity half-lives can well reproduce the experimental data. Moreover, we use this model to predict the 2p radioactivity half-lives of 22 candidates whose 2p radioactivity is energetically allowed or observed but not yet quantied in evaluated nuclear properties table NUBASE2016. The predicted results are in good agreement with the ones obtained by using Gamow-like model, effective liquid drop model (ELDM), generalized liquid drop model (GLDM) as well as a four-parameter formula.

nucl-th↗

arXiv2021

Dynamic Analysis of Nonlinear Civil Engineering Structures using Artificial Neural Network with Adaptive Training

Dynamic analysis of structures subjected to earthquake excitation is a time-consuming process, particularly in the case of extremely small time step required, or in the presence of high geometric and material nonlinearity. Performing parametric studies in such cases is even more tedious. The advancement of computer graphics hardware in recent years enables efficient training of artificial neural networks that are well-known to be capable of learning highly nonlinear mappings. In this study, artificial neural networks are developed with adaptive training algorithms, which enables automatic nodes generation and layers addition. The hyperbolic tangent function is selected as the activation function. Stochastic Gradient Descent and Back Propagation algorithms are adopted to train the networks. The neural networks initiate with a small number of hidden layers and nodes. During training, the performance of the network is continuously tracked, and new nodes or layers are added to the hidden layers if the neural network reaches its capacity. At the end of the training process, the network with appropriate architecture is automatically formed. The performance of the networks has been validated for inelastic shear frames, as well as rocking structures, of which both are first built in finite element program for dynamic analysis to generate training data. Results have shown the developed networks can successfully predict the time-history response of the shear frame and the rock structure subjected to real ground motion records. The efficiency of the proposed neural networks is also examined, which shows the computational time can be reduced by 43% by the neural networks method than FE models. This indicates the trained networks can be utilized to generate rocking spectrums of structures more efficiently which demands a large number of time-history analyses.

cs.LG↗

arXiv2021

Image-based monitoring of bolt loosening through deep-learning-based integrated detection and tracking

Structural bolts are critical components used in different structural elements, such as beam-column connections and friction damping devices. The clamping force in structural bolts is highly influenced by the bolt rotation. Much of the existing vision-based research about bolt rotation estimation relies on traditional computer vision algorithms such as Hough Transform to assess static images of bolts. This requires careful image preprocessing, and it may not perform well in the situation of complicated bolt assemblies, or in the presence of surrounding objects and background noise, thus hindering their real-world applications. In this study, an integrated real-time detect-track method, namely RTDT-Bolt, is proposed to monitor the bolt rotation angle. First, a real-time convolutional-neural-networks-based object detector, named YOLOv3-tiny, is established and trained to localize structural bolts. Then, the target-free object tracking algorithm based on optical flow is implemented, to continuously monitor and quantify the rotation of structural bolts. In order to enhance the tracking performance against background noise and potential illumination changes during tracking, the YOLOv3-tiny is integrated with the optical flow tracking algorithm to re-detect the bolts when the tracking gets lost. Extensive parameter studies were conducted to identify optimal tracking performance and examine the potential limitations. The results indicate the RTDT-Bolt method can greatly enhance the tracking performance of bolt rotation, which can achieve over 90% accuracy using the recommended range for the parameters.

cs.CV↗

arXiv2021

Postdisaster image-based damage detection and repair cost estimation of reinforced concrete buildings using dual convolutional neural networks

Reinforced concrete buildings are commonly used around the world. With recent earthquakes worldwide, rapid structural damage inspection and repair cost evaluation are crucial for building owners and policy makers to make informed risk management decisions. To improve the efficiency of such inspection, advanced computer vision techniques based on convolution neural networks have been adopted in recent research to rapidly quantify the damage state of structures. In this paper, an advanced object detection neural network, named YOLO-v2, is implemented which achieves 98.2% and 84.5% average precision in training and testing, respectively. The proposed YOLO-v2 is used in combination with the classification neural network, which improves the identification accuracy for critical damage state of reinforced concrete structures by 7.5%. The improved classification procedures allow engineers to rapidly and more accurately quantify the damage states of the structure, and also localize the critical damage features. The identified damage state can then be integrated with the state-of-the-art performance evaluation framework to quantify the financial losses of critical reinforced concrete buildings. The results can be used by the building owners and decision makers to make informed risk management decisions immediately after the strong earthquake shaking. Hence, resources can be allocated rapidly to improve the resiliency of the community.

cs.CV↗

arXiv2021

Systematic study of two-proton radioactivity half-lives within the two-potential approach with Skyrme-Hartree-Fock

In this work, we systematically study the two-proton($2p$) radioactivity half-lives using the two-potential approach while the nuclear potential is obtained by using Skyrme-Hartree-Fock approach with the Skyrme effective interaction of {SLy8}. For true $2p$ radioactivity($Q_{2p}$ $>$ 0 and $Q_p$ $< $0, where the $Q_p$ and $Q_{2p}$ are the released energy of the one-proton and two-proton radioactivity), the standard deviation between the experimental half-lives and our theoretical calculations is {0.701}. In addition, we extend this model to predict the half-lives of 15 possible $2p$ radioactivity candidates with $Q_{2p}$ $>$ 0 taken from the evaluated atomic mass table AME2016. The calculated results indicate that a clear linear relationship between the logarithmic $2p$ radioactivity half-lives $\rm{log}_{10}T_{1/2}$ and coulomb parameters [ ($Z_{d}^{0.8}$+$l^{0.25}$)$Q_{2p}^{-1/2}$] considered the effect of orbital angular momentum proposed by Liu $et$ $al$ [Chin. Phys. C \textbf{45}, 024108 (2021)] is also existed. For comparison, the generalized liquid drop model(GLDM), the effective liquid drop model(ELDM) and Gamow-like model are also used. Our predicted results are consistent with the ones obtained by the other models.

nucl-th↗

arXiv2021

Contrastive Learning for Many-to-many Multilingual Neural Machine Translation

Existing multilingual machine translation approaches mainly focus on English-centric directions, while the non-English directions still lag behind. In this work, we aim to build a many-to-many translation system with an emphasis on the quality of non-English language directions. Our intuition is based on the hypothesis that a universal cross-language representation leads to better multilingual translation performance. To this end, we propose mRASP2, a training method to obtain a single unified multilingual translation model. mRASP2 is empowered by two techniques: a) a contrastive learning scheme to close the gap among representations of different languages, and b) data augmentation on both multiple parallel and monolingual data to further align token representations. For English-centric directions, mRASP2 outperforms existing best unified model and achieves competitive or even better performance than the pre-trained and fine-tuned model mBART on tens of WMT's translation directions. For non-English directions, mRASP2 achieves an improvement of average 10+ BLEU compared with the multilingual Transformer baseline. Code, data and trained models are available at https://github.com/PANXiao1994/mRASP2.

cs.CL↗

arXiv2021

Pre-training Multilingual Neural Machine Translation by Leveraging Alignment Information

We investigate the following question for machine translation (MT): can we develop a single universal MT model to serve as the common seed and obtain derivative and improved models on arbitrary language pairs? We propose mRASP, an approach to pre-train a universal multilingual neural machine translation model. Our key idea in mRASP is its novel technique of random aligned substitution, which brings words and phrases with similar meanings across multiple languages closer in the representation space. We pre-train a mRASP model on 32 language pairs jointly with only public datasets. The model is then fine-tuned on downstream language pairs to obtain specialized MT models. We carry out extensive experiments on 42 translation directions across a diverse settings, including low, medium, rich resource, and as well as transferring to exotic language pairs. Experimental results demonstrate that mRASP achieves significant performance improvement compared to directly training on those target pairs. It is the first time to verify that multiple low-resource language pairs can be utilized to improve rich resource MT. Surprisingly, mRASP is even able to improve the translation quality on exotic languages that never occur in the pre-training corpus. Code, data, and pre-trained models are available at https://github.com/linzehui/mRASP.

cs.CL↗

arXiv2020

New Geiger-Nuttall law for two-proton radioactivity

In the present work, combining with the Geiger-Nuttall law, a two-parameter empirical formula is proposed to study the two-proton (2p) radioactivity. Using this formula, the calculated 2p radioactivity half-lives are in good agreement with the experimental data as well as the calculated ones obtained by Goncalves et al: ([Phys. Lett. B 774, 14 (2017)]) using the effective liquid drop model (ELDM), Sreeja et al: ([Eur. Phys. J. A 55, 33 (2019)]) using a four-parameter empirical formula and Cui et al: ([Phys. Rev. C 101: 014301 (2020)]) using a generalized liquid drop model (GLDM). In addition, this two-parameter empirical formula is extended to predict the half-lives of 22 possible 2p radioactivity candidates, whose the 2p radioactivity released energy Q2p>0, obtained from the latest evaluated atomic mass table AME2016. The predicted results have good consistency with ones using other theoretical models such as the ELDM, GLDM and four-parameter empirical formula.

nucl-th↗

arXiv2020

The Volctrans Machine Translation System for WMT20

This paper describes our VolcTrans system on WMT20 shared news translation task. We participated in 8 translation directions. Our basic systems are based on Transformer, with several variants (wider or deeper Transformers, dynamic convolutions). The final system includes text pre-process, data selection, synthetic data generation, advanced model ensemble, and multilingual pre-training.

cs.CL↗

arXiv2020

Systematic study of proton radioactivity of spherical proton emitters with Skyrme interactions

Proton radioactivity is an important and common process of the natural radioactivity of proton-rich nuclei. In our previous work [J. H. Cheng et al., Nucl. Rhys. A 997, 121717 (2020)], we systematically studied the proton radioactivity half-lives of 53< Z <83 nuclei within the two-potential approach with Skyrme-Hartree-Fock. The calculations can well reproduce the experimental data of spherical nuclei. The present work is an extension of the previous work. We systematically study 32 proton radioactivity half-lives of spherical nuclei with the two-potential approach with Skyrme-Hartree-Fock (TPA-SHF) within 115 different versions of the Skyrme interactions. The calculated results indicate that the SLy7 Skyrme interaction gives the lowest r.m.s. deviation in the description of the experimental data of the spherical proton emitters among all the different versions of the Skyrme interactions. In addition, we use this model with the SLy7 Skyrme interaction to predict the proton radioactivity half-lives of 7 spherical nuclei in the same region, whose proton radioactivity is energetically allowed or has been observed but is not yet quantified.

nucl-th↗

arXiv2016

Bicycle-Sharing System Analysis and Trip Prediction

Bicycle-sharing systems, which can provide shared bike usage services for the public, have been launched in many big cities. In bicycle-sharing systems, people can borrow and return bikes at any stations in the service region very conveniently. Therefore, bicycle-sharing systems are normally used as a short-distance trip supplement for private vehicles as well as regular public transportation. Meanwhile, for stations located at different places in the service region, the bike usages can be quite skewed and imbalanced. Some stations have too many incoming bikes and get jammed without enough docks for upcoming bikes, while some other stations get empty quickly and lack enough bikes for people to check out. Therefore, inferring the potential destinations and arriving time of each individual trip beforehand can effectively help the service providers schedule manual bike re-dispatch in advance. In this paper, we will study the individual trip prediction problem for bicycle-sharing systems. To address the problem, we study a real-world bicycle-sharing system and analyze individuals' bike usage behaviors first. Based on the analysis results, a new trip destination prediction and trip duration inference model will be introduced. Experiments conducted on a real-world bicycle-sharing system demonstrate the effectiveness of the proposed model.

cs.CY↗