arXiv subjects

Chen Xu

Publications and source records attributed to Chen Xu.

At least 199 records · Page 11Linked to original sources

arXiv2022

Solar Radiation Ramping Events Modeling Using Spatio-temporal Point Processes

Modeling and predicting solar events, particularly the solar ramping event, is critical for improving situational awareness for solar power generation systems. It has been acknowledged that weather conditions such as temperature, humidity, and cloud density can significantly impact the emergence and position of solar ramping events. As a result, modeling these events with complex spatio-temporal correlations is highly challenging. To tackle the question, we adopt a novel spatio-temporal categorical point process model, which intuitively and effectively addresses correlation and interaction among ramping events. We demonstrate the interpretability and predictive power of our model on extensive real-data experiments.

stat.AP↗

arXiv2022

Conformal prediction set for time-series

When building either prediction intervals for regression (with real-valued response) or prediction sets for classification (with categorical responses), uncertainty quantification is essential to studying complex machine learning methods. In this paper, we develop Ensemble Regularized Adaptive Prediction Set (ERAPS) to construct prediction sets for time-series (with categorical responses), based on the prior work of [Xu and Xie, 2021]. In particular, we allow unknown dependencies to exist within features and responses that arrive in sequence. Method-wise, ERAPS is a distribution-free and ensemble-based framework that is applicable for arbitrary classifiers. Theoretically, we bound the coverage gap without assuming data exchangeability and show asymptotic set convergence. Empirically, we demonstrate valid marginal and conditional coverage by ERAPS, which also tends to yield smaller prediction sets than competing methods.

stat.ML↗

arXiv2022

COSPLAY: Concept Set Guided Personalized Dialogue Generation Across Both Party Personas

Maintaining a consistent persona is essential for building a human-like conversational model. However, the lack of attention to the partner makes the model more egocentric: they tend to show their persona by all means such as twisting the topic stiffly, pulling the conversation to their own interests regardless, and rambling their persona with little curiosity to the partner. In this work, we propose COSPLAY(COncept Set guided PersonaLized dialogue generation Across both partY personas) that considers both parties as a "team": expressing self-persona while keeping curiosity toward the partner, leading responses around mutual personas, and finding the common ground. Specifically, we first represent self-persona, partner persona and mutual dialogue all in the concept sets. Then, we propose the Concept Set framework with a suite of knowledge-enhanced operations to process them such as set algebras, set expansion, and set distance. Based on these operations as medium, we train the model by utilizing 1) concepts of both party personas, 2) concept relationship between them, and 3) their relationship to the future dialogue. Extensive experiments on a large public dataset, Persona-Chat, demonstrate that our model outperforms state-of-the-art baselines for generating less egocentric, more human-like, and higher quality responses in both automatic and human evaluations.

cs.CL↗

arXiv2022

Learnable Irrelevant Modality Dropout for Multimodal Action Recognition on Modality-Specific Annotated Videos

With the assumption that a video dataset is multimodality annotated in which auditory and visual modalities both are labeled or class-relevant, current multimodal methods apply modality fusion or cross-modality attention. However, effectively leveraging the audio modality in vision-specific annotated videos for action recognition is of particular challenge. To tackle this challenge, we propose a novel audio-visual framework that effectively leverages the audio modality in any solely vision-specific annotated dataset. We adopt the language models (e.g., BERT) to build a semantic audio-video label dictionary (SAVLD) that maps each video label to its most K-relevant audio labels in which SAVLD serves as a bridge between audio and video datasets. Then, SAVLD along with a pretrained audio multi-label model are used to estimate the audio-visual modality relevance during the training phase. Accordingly, a novel learnable irrelevant modality dropout (IMD) is proposed to completely drop out the irrelevant audio modality and fuse only the relevant modalities. Moreover, we present a new two-stream video Transformer for efficiently modeling the visual modalities. Results on several vision-specific annotated datasets including Kinetics400 and UCF-101 validated our framework as it outperforms most relevant action recognition methods.

cs.CV↗

arXiv2022

Preventing the Spread of Online Harms: Physics of Contagion across Multi-Platform Social Media and Metaverses

We present a minimal yet empirically-grounded theory for the spread of online harms (e.g. misinformation, hate) across current multi-platform social media and future Metaverses. New physics emerges from the interplay between the intrinsic heterogeneity among online communities and platforms, their clustering dynamics generated through user-created links and sudden moderator shutdowns, and the contagion process. The theory provides an online `R-nought' criterion to prevent system-wide spreading; it predicts re-entrant spreading phases; it establishes the level of digital vaccination required for online herd immunity; and it can be applied at multiple scales.

physics.soc-ph↗

arXiv2022

SMLE: An R Package for Joint Feature Screening in Ultrahigh-dimensional GLMs

The sparsity-restricted maximum likelihood estimator (SMLE) has received considerable attention for feature screening in ultrahigh-dimensional regression. SMLE is a computationally convenient method that naturally incorporates the joint effects among features in the screening process. We develop a publicly available R package SMLE, which provides a user-friendly environment to carry out SMLE in generalized linear models. In particular, the package includes functions to conduct SMLE-screening and post-screening selection using SMLE with popular selection criteria such as AIC and (extended) BIC. The package gives users the flexibility in controlling a series of screening parameters and accommodates both numerical and categorical feature input. The usage of the package is illustrated on extensive numerical examples, where the promising performance of SMLE is well observed.

stat.OT↗

arXiv2021

Electrostatic Discovery Atomic Force Microscopy

While offering unprecedented resolution of atomic and electronic structure, Scanning Probe Microscopy techniques have found greater challenges in providing reliable electrostatic characterization at the same scale. In this work, we introduce Electrostatic Discovery Atomic Force Microscopy, a machine learning based method which provides immediate quantitative maps of the electrostatic potential directly from Atomic Force Microscopy images with functionalized tips. We apply this to characterize the electrostatic properties of a variety of molecular systems and compare directly to reference simulations, demonstrating good agreement. This approach opens the door to reliable atomic scale electrostatic maps on any system with minimal computational overhead.

cond-mat.mes-hall↗

arXiv2021

Stabilization of continuous-time Markov/semi-Markov jump linear systems via finite data-rate feedback

This paper investigates almost sure exponential stabilization of continuous-time Markov jump linear systems (MJLSs) under communication data-rate constraints by introducing sampling and quantization into the feedback control. Different from previous works, the sampling times and the jump times are independent of each other in this paper. The quantization is recursively adjusted on the sampling time, and its updating strategy does not depend on the switching in a sampling interval. In other words, the explicit value of the switching signal in a sampling interval is not necessary. The numerically testable condition is developed to ensure almost sure exponential stabilization of MJLSs under the proposed communication and control protocols. We also drop the assumption of stabilizability of all individual modes required in previous works about the switched systems. Moreover, we extend the result to the case of continuous-time semi-Markov jump linear systems (semi-MJLSs) via the semi-Markov kernel approach. Finally, some numerical examples are presented to illustrate the effectiveness of the proposed communication and control protocols.

math.DS↗

arXiv2021

DPR-CAE: Capsule Autoencoder with Dynamic Part Representation for Image Parsing

Parsing an image into a hierarchy of objects, parts, and relations is important and also challenging in many computer vision tasks. This paper proposes a simple and effective capsule autoencoder to address this issue, called DPR-CAE. In our approach, the encoder parses the input into a set of part capsules, including pose, intensity, and dynamic vector. The decoder introduces a novel dynamic part representation (DPR) by combining the dynamic vector and a shared template bank. These part representations are then regulated by corresponding capsules to composite the final output in an interpretable way. Besides, an extra translation-invariant module is proposed to avoid directly learning the uncertain scene-part relationship in our DPR-CAE, which makes the resulting method achieves a promising performance gain on $rm$-MNIST and $rm$-Fashion-MNIST. % to model the scene-object relationship DPR-CAE can be easily combined with the existing stacked capsule autoencoder and experimental results show it significantly improves performance in terms of unsupervised object classification. Our code is available in the Appendix.

cs.CV↗

arXiv2021

A Credibility-aware Swarm-Federated Deep Learning Framework in Internet of Vehicles

Federated Deep Learning (FDL) is helping to realize distributed machine learning in the Internet of Vehicles (IoV). However, FDL's global model needs multiple clients to upload learning model parameters, thus still existing unavoidable communication overhead and data privacy risks. The recently proposed Swarm Learning (SL) provides a decentralized machine-learning approach uniting edge computing and blockchain-based coordination without the need for a central coordinator. This paper proposes a Swarm-Federated Deep Learning framework in the IoV system (IoV-SFDL) that integrates SL into the FDL framework. The IoV-SFDL organizes vehicles to generate local SL models with adjacent vehicles based on the blockchain empowered SL, then aggregates the global FDL model among different SL groups with a proposed credibility weights prediction algorithm. Extensive experimental results demonstrate that compared with the baseline frameworks, the proposed IoV-SFDL framework achieves a 16.72% reduction in edge-to-global communication overhead while improving about 5.02% in model performance with the same training iterations.

eess.SY↗

arXiv2021

Reactivity of transition-metal alloys to oxygen and sulphur

Oxidation and tarnishing are the two most common initial steps in the corrosive process of metals at ambient conditions. These are always initiated with O and S binding to a metallic surface, so that one can use the binding energy as a rough proxy for the metal reactivity. With this in mind, we present a systematic study of the binding energy of O and S across the entire transition-metals composition space, namely we explore the binding energy of {\bf 88} single-phase transition metals and of {\bf 646} transition-metal binary alloys. The analysis is performed by defining a suitable descriptor for the binding energy. This is here obtained by fitting several schemes, based on the original Newns-Anderson model, against density-functional-theory data for the 4$d$ transition metal series. Such descriptor is then applied to a vast database of electronic structures of transition-metal alloys, for which we are able to predict the range of binding energies across both the compositional and the structural space. Finally, we extend our analysis to ternary transition-metal alloys and identify the most resilient compounds to O and S binding.

cond-mat.mtrl-sci↗

arXiv2021

The NiuTrans End-to-End Speech Translation System for IWSLT 2021 Offline Task

This paper describes the submission of the NiuTrans end-to-end speech translation system for the IWSLT 2021 offline task, which translates from the English audio to German text directly without intermediate transcription. We use the Transformer-based model architecture and enhance it by Conformer, relative position encoding, and stacked acoustic and textual encoding. To augment the training data, the English transcriptions are translated to German translations. Finally, we employ ensemble decoding to integrate the predictions from several models trained with the different datasets. Combining these techniques, we achieve 33.84 BLEU points on the MuST-C En-De test set, which shows the enormous potential of the end-to-end model.

cs.CL↗

arXiv2021

A new mathematical model for dispersion of Rayleigh wave and a machine learning based inversion solver

In this work, by introducing the seismic impedance tensor we propose a new Rayleigh wave dispersion function in a homogeneous and layered medium of the Earth, which provides an efficient way to compute the dispersion curve -- a relation between the frequencies and the phase velocities. With this newly established forward model, based on the Mixture Density Networks (MDN) we develop a machine learning based inversion approach, named as FW-MDN, for the problem of estimating the S-wave velocity from the dispersion curves. The method FW-MDN deals with the non-uniqueness issue encountered in studies that invert dispersion curves for crust and upper mantle models and attains a satisfactory performance on the dataset with various noise structure. Numerical simulations are performed to show that the FW-MDN possesses the characteristics of easy calculation, efficient computation, and high precision for the model characterization.

math.NA↗

arXiv2021

Stacked Acoustic-and-Textual Encoding: Integrating the Pre-trained Models into Speech Translation Encoders

Encoder pre-training is promising in end-to-end Speech Translation (ST), given the fact that speech-to-translation data is scarce. But ST encoders are not simple instances of Automatic Speech Recognition (ASR) or Machine Translation (MT) encoders. For example, we find that ASR encoders lack the global context representation, which is necessary for translation, whereas MT encoders are not designed to deal with long but locally attentive acoustic sequences. In this work, we propose a Stacked Acoustic-and-Textual Encoding (SATE) method for speech translation. Our encoder begins with processing the acoustic sequence as usual, but later behaves more like an MT encoder for a global representation of the input sequence. In this way, it is straightforward to incorporate the pre-trained models into the system. Also, we develop an adaptor module to alleviate the representation inconsistency between the pre-trained ASR encoder and MT encoder, and develop a multi-teacher knowledge distillation method to preserve the pre-training knowledge. Experimental results on the LibriSpeech En-Fr and MuST-C En-De ST tasks show that our method achieves state-of-the-art BLEU scores of 18.3 and 25.2. To our knowledge, we are the first to develop an end-to-end ST system that achieves comparable or even better BLEU performance than the cascaded ST counterpart when large-scale ASR and MT data is available.

cs.CL↗

arXiv2021

Conformal Anomaly Detection on Spatio-Temporal Observations with Missing Data

We develop a distribution-free, unsupervised anomaly detection method called ECAD, which wraps around any regression algorithm and sequentially detects anomalies. Rooted in conformal prediction, ECAD does not require data exchangeability but approximately controls the Type-I error when data are normal. Computationally, it involves no data-splitting and efficiently trains ensemble predictors to increase statistical power. We demonstrate the superior performance of ECAD on detecting anomalous spatio-temporal traffic flow.

stat.AP↗

arXiv2021

Stable Nonlinear and IQ Imbalance RF Fingerprint for Wireless OFDM Devices

An estimation method of Radio Frequency fingerprint (RFF) based on the physical hardware properties of the nonlinearity and in-phase and quadrature (IQ) imbalance of the transmitter is proposed for the authentication of wireless orthogonal frequency division multiplexing (OFDM) devices. Firstly, the parameters of the nonlinearity of the transmitter and finite impulse response (FIR) of the wireless multipath channel are estimated with a Hammerstein system parameter separation technique. Secondly, the best IQ imbalance parameter combination estimation is obtained with a searching algorithm and the applied conjugate anti-symmetric pilots. Thirdly, the estimations of the nonlinear coefficients and IQ imbalance parameter combination are considered as a novel RFF, the features are extracted from the novel RFF, and a k-Nearest Neighbor (k-NN) classifier is used to classify the communication devices with the features. It is demonstrated with the numerical experiments of five transmitters that the novel RFF eliminates the adverse effect of the wireless channel and is therefore stable. The proposed RF fingerprinting method is helpful for the high-strength authentication of the OFDM communication devices with subtle differences from the same model and same series.

cs.IT↗

arXiv2021

MMA Regularization: Decorrelating Weights of Neural Networks by Maximizing the Minimal Angles

The strong correlation between neurons or filters can significantly weaken the generalization ability of neural networks. Inspired by the well-known Tammes problem, we propose a novel diversity regularization method to address this issue, which makes the normalized weight vectors of neurons or filters distributed on a hypersphere as uniformly as possible, through maximizing the minimal pairwise angles (MMA). This method can easily exert its effect by plugging the MMA regularization term into the loss function with negligible computational overhead. The MMA regularization is simple, efficient, and effective. Therefore, it can be used as a basic regularization method in neural network training. Extensive experiments demonstrate that MMA regularization is able to enhance the generalization ability of various modern models and achieves considerable performance improvements on CIFAR100 and TinyImageNet datasets. In addition, experiments on face verification show that MMA regularization is also effective for feature learning. Code is available at: https://github.com/wznpub/MMA_Regularization.

cs.LG↗

arXiv2021

Smart Scheduling based on Deep Reinforcement Learning for Cellular Networks

To improve the system performance towards the Shannon limit, advanced radio resource management mechanisms play a fundamental role. In particular, scheduling should receive much attention, because it allocates radio resources among different users in terms of their channel conditions and QoS requirements. The difficulties of scheduling algorithms are the tradeoffs need to be made among multiple objectives, such as throughput, fairness and packet drop rate. We propose a smart scheduling scheme based on deep reinforcement learning (DRL). We not only verify the performance gain achieved, but also provide implementation-friend designs, i.e., a scalable neural network design for the agent and a virtual environment training framework. With the scalable neural network design, the DRL agent can easily handle the cases when the number of active users is time-varying without the need to redesign and retrain the DRL agent. Training the DRL agent in a virtual environment offline first and using it as the initial version in the practical usage helps to prevent the system from suffering from performance and robustness degradation due to the time-consuming training. Through both simulations and field tests, we show that the DRL-based smart scheduling outperforms the conventional scheduling method and can be adopted in practical systems.

cs.LG↗