arXiv subjects

Cheng Gong

Publications and source records attributed to Cheng Gong.

At least 55 records · Page 3Linked to original sources

arXiv2024

An Initial Investigation of Language Adaptation for TTS Systems under Low-resource Scenarios

Self-supervised learning (SSL) representations from massively multilingual models offer a promising solution for low-resource language speech tasks. Despite advancements, language adaptation in TTS systems remains an open problem. This paper explores the language adaptation capability of ZMM-TTS, a recent SSL-based multilingual TTS system proposed in our previous work. We conducted experiments on 12 languages using limited data with various fine-tuning configurations. We demonstrate that the similarity in phonetics between the pre-training and target languages, as well as the language category, affects the target language's adaptation performance. Additionally, we find that the fine-tuning dataset size and number of speakers influence adaptability. Surprisingly, we also observed that using paired data for fine-tuning is not always optimal compared to audio-only data. Beyond speech intelligibility, our analysis covers speaker similarity, language identification, and predicted MOS.

cs.CL↗

arXiv2024

Motion planning for off-road autonomous driving based on human-like cognition and weight adaptation

Driving in an off-road environment is challenging for autonomous vehicles due to the complex and varied terrain. To ensure stable and efficient travel, the vehicle requires consideration and balancing of environmental factors, such as undulations, roughness, and obstacles, to generate optimal trajectories that can adapt to changing scenarios. However, traditional motion planners often utilize a fixed cost function for trajectory optimization, making it difficult to adapt to different driving strategies in challenging irregular terrains and uncommon scenarios. To address these issues, we propose an adaptive motion planner based on human-like cognition and cost evaluation for off-road driving. First, we construct a multi-layer map describing different features of off-road terrains, including terrain elevation, roughness, obstacle, and artificial potential field map. Subsequently, we employ a CNN-LSTM network to learn the trajectories planned by human drivers in various off-road scenarios. Then, based on human-like generated trajectories in different environments, we design a primitive-based trajectory planner that aims to mimic human trajectories and cost weight selection, generating trajectories that are consistent with the dynamics of off-road vehicles. Finally, we compute optimal cost weights and select and extend behavioral primitives to generate highly adaptive, stable, and efficient trajectories. We validate the effectiveness of the proposed method through experiments in a desert off-road environment with complex terrain and varying road conditions. The experimental results show that the proposed human-like motion planner has excellent adaptability to different off-road conditions. It shows real-time operation, greater stability, and more human-like planning ability in diverse and challenging scenarios.

cs.RO↗

arXiv2024

Deformations of Zappatic stable surfaces and their Galois covers

This paper considers some algebraic surfaces that can deform to planar Zappatic stable surfaces with a unique singularity of type En. We prove that the Galois covers of these surfaces are all simply connected of general type, for n >= 4, and we give a formula for Chern numbers of such Galois covers. As an application, we prove that such surfaces do not exist for n>30. Furthermore, Kollar improves the result to n>9 in Appendix 5.

math.AG↗

arXiv2024

Minimize Quantization Output Error with Bias Compensation

Quantization is a promising method that reduces memory usage and computational intensity of Deep Neural Networks (DNNs), but it often leads to significant output error that hinder model deployment. In this paper, we propose Bias Compensation (BC) to minimize the output error, thus realizing ultra-low-precision quantization without model fine-tuning. Instead of optimizing the non-convex quantization process as in most previous methods, the proposed BC bypasses the step to directly minimize the quantizing output error by identifying a bias vector for compensation. We have established that the minimization of output error through BC is a convex problem and provides an efficient strategy to procure optimal solutions associated with minimal output error,without the need for training or fine-tuning. We conduct extensive experiments on Vision Transformer models and Large Language Models, and the results show that our method notably reduces quantization output error, thereby permitting ultra-low-precision post-training quantization and enhancing the task performance of models. Especially, BC improves the accuracy of ViT-B with 4-bit PTQ4ViT by 36.89% on the ImageNet-1k task, and decreases the perplexity of OPT-350M with 3-bit GPTQ by 5.97 on WikiText2.The code is in https://github.com/GongCheng1919/bias-compensation.

cs.CV↗

arXiv2024

On Galois covers of a union of Zappatic surfaces of type $R_k$

We investigate the topological structures of Galois covers of a union of two Zappatic surfaces of type $R_k$. We prove that the Galois covers of such surfaces are simply-connected surfaces of general type. We also compute their Chern numbers and topological indices.

math.AG↗

arXiv2023

Two-Step Electronic Response to Magnetic Ordering in a van der Waals Ferromagnet

The two-dimensional (2D) material Cr$_2$Ge$_2$Te$_6$ is a member of the class of insulating van der Waals magnets. Here, using high resolution angle-resolved photoemission spectroscopy in a detailed temperature dependence study, we identify a clear response of the electronic structure to a dimensional crossover in the form of two distinct temperature scales marking onsets of modifications in the electronic structure. Specifically, we observe Te $p$-orbital-dominated bands to undergo changes at the Curie transition temperature T$_C$ while the Cr $d$-orbital-dominated bands begin evolving at a higher temperature scale. Combined with neutron scattering, density functional theory calculations, and Monte Carlo simulations, we find that the electronic system can be consistently understood to respond sequentially to the distinct temperatures at which in-plane and out-of-plane spin correlations exceed a characteristic length scale. Our findings reveal the sensitivity of the orbital-selective electronic structure for probing the dynamical evolution of local moment correlations in vdW insulating magnets.

cond-mat.str-el↗

arXiv2023

On the Galois covers of degenerations of surfaces of minimal degree

We investigate the topological structures of Galois covers of surfaces of minimal degree (i.e., degree n) in n+1 dimensional complex projective space. We prove that for n is greater than or equal to 5, the Galois covers of any surfaces of minimal degree are simply-connected surfaces of general type.

math.AG↗

arXiv2023

Terahertz reconfigurable multi-functional metamaterials based on 3D printed mortise-tenon structures

The emergence of metamaterial has provided an unprecedented ability to manipulate electromagnetic waves, especially in the terahertz band where there is a lack of natural response materials. However, most metamaterials are fixed single function due to the fixed structure at the beginning of design. The paper reports a reconfigurable multi-functional terahertz metamaterial with variable structures based on mortise and tenon mechanism. And a hybrid 3D printing method based on FDM and E-jet is proposed to fabricate the metamaterials, which simplifies the processing process, improves the speed, and reduces the cost compared to traditional semiconductor processing methods. Through flexible mortise and tenon connections, the metamaterial can achieve: (1) narrowband transmission and broadband absorption; (2) perfect reflection; (3) narrowband reflection and broadband absorption. Relying on ingenious design and processing, the multi-functional metamaterials are expected to be widely used in fields such as electromagnetic shielding, radar stealth, communication and so on.

physics.optics↗

arXiv2023

Nanotips for 0.5THz scattering scanning near field microscopy

This manuscript demonstrates the theory, design, and simulation of scattering scanning near field microscopy (s-SNOM) in 0.5THz. A comprehensive simulation model of nanotips' geometry, sample materials, and incident field is established to significantly improve the scattering efficiency and spatial resolution to achieve optimal performance. The theoretical model is based on full-wave simulation and dipole moment analysis which can describe the overall nanotip's geometry information to screen the optimal parameters corresponding to the 0.5THz, which is the center frequency of most THz sources. The customized nanotip can achieve 40 nm $(λ/ 15000)$ spatial resolution while maintaining an excellent scattering efficiency. This nanotip design method doesn't depend on the homogenization commercial AFM tips, providing an approach for customized nanotip design of THz wave scattering near field microscopy.

physics.optics↗

arXiv2023

AutoQNN: An End-to-End Framework for Automatically Quantizing Neural Networks

Exploring the expected quantizing scheme with suitable mixed-precision policy is the key point to compress deep neural networks (DNNs) in high efficiency and accuracy. This exploration implies heavy workloads for domain experts, and an automatic compression method is needed. However, the huge search space of the automatic method introduces plenty of computing budgets that make the automatic process challenging to be applied in real scenarios. In this paper, we propose an end-to-end framework named AutoQNN, for automatically quantizing different layers utilizing different schemes and bitwidths without any human labor. AutoQNN can seek desirable quantizing schemes and mixed-precision policies for mainstream DNN models efficiently by involving three techniques: quantizing scheme search (QSS), quantizing precision learning (QPL), and quantized architecture generation (QAG). QSS introduces five quantizing schemes and defines three new schemes as a candidate set for scheme search, and then uses the differentiable neural architecture search (DNAS) algorithm to seek the layer- or model-desired scheme from the set. QPL is the first method to learn mixed-precision policies by reparameterizing the bitwidths of quantizing schemes, to the best of our knowledge. QPL optimizes both classification loss and precision loss of DNNs efficiently and obtains the relatively optimal mixed-precision model within limited model size and memory footprint. QAG is designed to convert arbitrary architectures into corresponding quantized ones without manual intervention, to facilitate end-to-end neural network quantization. We have implemented AutoQNN and integrated it into Keras. Extensive experiments demonstrate that AutoQNN can consistently outperform state-of-the-art quantization.

cs.LG↗

arXiv2023

Gravitational wave constraints on non-birefringent dispersions of gravitational waves due to Lorentz violations with GWTC-3

The standard model extension (SME) is an effective field theory framework that can be used to study the possible violations of Lorentz symmetry in the gravitational interaction. In the SME's gauge invariant linearized gravity sector, the dispersion relation of GWs is modified, resulting in anisotropy, birefringence, and dispersion effects in the propagation of GWs. In this paper, we mainly focus on the non-birefringent and anisotropic dispersion relation in the propagation of GWs due to the violation of Lorentz symmetry. With the modified dispersion relation, we calculate the corresponding modified waveform of GWs generated by the coalescence of compact binaries. We consider the effects from the operators with the lowest mass dimension $d=6$ in the gauge invariant linearized gravity sector of the SME which are expected to have the dominant Lorentz-violating effect on the propagation of GWs. For this case, the Lorentz-violating effects are presented by 25 coefficients and we constrain them independently by the ``maximal-reach" approach. We use 90 high-confidence GW events in the GWTC-3 catalog and use {\tt Bilby}, an open source software, and {\tt Dynest}, a nested sampling package, to perform parameter estimation with the modified waveform. We do not find any evidence of Lorentz violation in the GWs data and give a $90\%$ confidence interval for each Lorentz violating coefficient.

gr-qc↗

arXiv2022

Continual Interactive Behavior Learning With Traffic Divergence Measurement: A Dynamic Gradient Scenario Memory Approach

Developing autonomous vehicles (AVs) helps improve the road safety and traffic efficiency of intelligent transportation systems (ITS). Accurately predicting the trajectories of traffic participants is essential to the decision-making and motion planning of AVs in interactive scenarios. Recently, learning-based trajectory predictors have shown state-of-the-art performance in highway or urban areas. However, most existing learning-based models trained with fixed datasets may perform poorly in continuously changing scenarios. Specifically, they may not perform well in learned scenarios after learning the new one. This phenomenon is called "catastrophic forgetting". Few studies investigate trajectory predictions in continuous scenarios, where catastrophic forgetting may happen. To handle this problem, first, a novel continual learning (CL) approach for vehicle trajectory prediction is proposed in this paper. Then, inspired by brain science, a dynamic memory mechanism is developed by utilizing the measurement of traffic divergence between scenarios, which balances the performance and training efficiency of the proposed CL approach. Finally, datasets collected from different locations are used to design continual training and testing methods in experiments. Experimental results show that the proposed approach achieves consistently high prediction accuracy in continuous scenarios without re-training, which mitigates catastrophic forgetting compared to non-CL approaches. The implementation of the proposed approach is publicly available at https://github.com/BIT-Jack/D-GSM

cs.RO↗

arXiv2022

An Ensemble Learning Framework for Vehicle Trajectory Prediction in Interactive Scenarios

Precisely modeling interactions and accurately predicting trajectories of surrounding vehicles are essential to the decision-making and path-planning of intelligent vehicles. This paper proposes a novel framework based on ensemble learning to improve the performance of trajectory predictions in interactive scenarios. The framework is termed Interactive Ensemble Trajectory Predictor (IETP). IETP assembles interaction-aware trajectory predictors as base learners to build an ensemble learner. Firstly, each base learner in IETP observes historical trajectories of vehicles in the scene. Then each base learner handles interactions between vehicles to predict trajectories. Finally, an ensemble learner is built to predict trajectories by applying two ensemble strategies on the predictions from all base learners. Predictions generated by the ensemble learner are final outputs of IETP. In this study, three experiments using different data are conducted based on the NGSIM dataset. Experimental results show that IETP improves the predicting accuracy and decreases the variance of errors compared to base learners. In addition, IETP exceeds baseline models with 50% of the training data, indicating that IETP is data-efficient. Moreover, the implementation of IETP is publicly available at https://github.com/BIT-Jack/IETP.

cs.RO↗

arXiv2022

Gravitational wave constraints on Lorentz and parity violations in gravity: high-order spatial derivative cases

High-order spatial derivatives are of crucial importance for constructing the low energy effective action of a Lorentz or parity violating theory of quantum gravity. One example is the Hořava-Lifshitz gravity, in one has to consider at least the sixth-order spatial derivatives in the gravitational action, in order to make the theory power-counting renormalizable. In this paper, we consider the Lorentz and parity violating effects on the propagation of GWs due to the fifth and sixth-order spatial derivatives respectively. For this purpose we calculate the corresponding Lorentz and parity violating waveforms of GWs produced by the coalescence of compact binaries. By using these modified waveforms, we perform the full Bayesian inference with the help of the open source software \texttt{Bilby} on the selected GW events of binary black hole (BBH) and binary neutron stars (BNS) merges in the LIGO-Virgo catalogs GWTC-1 and GWTC-2. Overall we do not find any significant evidence of Lorentz and parity violation due to the fifth and sixth-order spatial derivatives and thus place lower bounds on the energy scales $M_{\rm LV} > 2.4 \times 10^{-16} \; {\rm GeV}$ for Lorentz violation and $M_{\rm PV} > 1.0 \times 10^{-14} \; {\rm GeV}$ for parity violation at 90\% confidence level. Both constraints represent the first constraints on the fifth- and sixth-order spatial derivative terms respectively in the framework of spatial covariant gravity by using the observational data of GWs.

gr-qc↗

arXiv2021

Elastic Significant Bit Quantization and Acceleration for Deep Neural Networks

Quantization has been proven to be a vital method for improving the inference efficiency of deep neural networks (DNNs). However, it is still challenging to strike a good balance between accuracy and efficiency while quantizing DNN weights or activation values from high-precision formats to their quantized counterparts. We propose a new method called elastic significant bit quantization (ESB) that controls the number of significant bits of quantized values to obtain better inference accuracy with fewer resources. We design a unified mathematical formula to constrain the quantized values of the ESB with a flexible number of significant bits. We also introduce a distribution difference aligner (DDA) to quantitatively align the distributions between the full-precision weight or activation values and quantized values. Consequently, ESB is suitable for various bell-shaped distributions of weights and activation of DNNs, thus maintaining a high inference accuracy. Benefitting from fewer significant bits of quantized values, ESB can reduce the multiplication complexity. We implement ESB as an accelerator and quantitatively evaluate its efficiency on FPGAs. Extensive experimental results illustrate that ESB quantization consistently outperforms state-of-the-art methods and achieves average accuracy improvements of 4.78%, 1.92%, and 3.56% over AlexNet, ResNet18, and MobileNetV2, respectively. Furthermore, ESB as an accelerator can achieve 10.95 GOPS peak performance of 1k LUTs without DSPs on the Xilinx ZCU102 FPGA platform. Compared with CPU, GPU, and state-of-the-art accelerators on FPGAs, the ESB accelerator can improve the energy efficiency by up to 65x, 11x, and 26x, respectively.

cs.CV↗

arXiv2021

Driver-Specific Risk Recognition in Interactive Driving Scenarios using Graph Representation

This paper presents a driver-specific risk recognition framework for autonomous vehicles that can extract inter-vehicle interactions. This extraction is carried out for urban driving scenarios in a driver-cognitive manner to improve the recognition accuracy of risky scenes. First, clustering analysis is applied to the operation data of drivers for learning the subjective assessment of risky scenes of different drivers and generating the corresponding risk label for each scene. Second, the graph representation model (GRM) is adopted to unify and construct the features of dynamic vehicles, inter-vehicle interactions and static traffic markings in real driving scenes into graphs. The driver-specific risk label provides ground truth to capture the risk evaluation criteria of different drivers. In addition, the graph model represents multiple features of the driving scenes. Therefore, the proposed framework can learn the risk-evaluating pattern of driving scenes of different drivers and establish driver-specific risk identifiers. Last, the performance of the proposed framework is evaluated via experiments conducted using real-world urban driving datasets collected by multiple drivers. The results show that the risks and their levels in real driving environments can be accurately recognized by the proposed framework.

cs.RO↗

arXiv2021

Prediction of Pedestrian Spatiotemporal Risk Levels for Intelligent Vehicles: A Data-driven Approach

In recent years, road safety has attracted significant attention from researchers and practitioners in the intelligent transport systems domain. As one of the most common and vulnerable groups of road users, pedestrians cause great concerns due to their unpredictable behavior and movement, as subtle misunderstandings in vehicle-pedestrian interaction can easily lead to risky situations or collisions. Existing methods use either predefined collision-based models or human-labeling approaches to estimate the pedestrians' risks. These approaches are usually limited by their poor generalization ability and lack of consideration of interactions between the ego vehicle and a pedestrian. This work tackles the listed problems by proposing a Pedestrian Risk Level Prediction system. The system consists of three modules. Firstly, vehicle-perspective pedestrian data are collected. Since the data contains information regarding the movement of both the ego vehicle and pedestrian, it can simplify the prediction of spatiotemporal features in an interaction-aware fashion. Using the long short-term memory model, the pedestrian trajectory prediction module predicts their spatiotemporal features in the subsequent five frames. As the predicted trajectory follows certain interaction and risk patterns, a hybrid clustering and classification method is adopted to explore the risk patterns in the spatiotemporal features and train a risk level classifier using the learned patterns. Upon predicting the spatiotemporal features of pedestrians and identifying the corresponding risk level, the risk patterns between the ego vehicle and pedestrians are determined. Experimental results verified the capability of the PRLP system to predict the risk level of pedestrians, thus supporting the collision risk assessment of intelligent vehicles and providing safety warnings to both vehicles and pedestrians.

cs.RO↗

arXiv2021

Using multiple reference audios and style embedding constraints for speech synthesis

The end-to-end speech synthesis model can directly take an utterance as reference audio, and generate speech from the text with prosody and speaker characteristics similar to the reference audio. However, an appropriate acoustic embedding must be manually selected during inference. Due to the fact that only the matched text and speech are used in the training process, using unmatched text and speech for inference would cause the model to synthesize speech with low content quality. In this study, we propose to mitigate these two problems by using multiple reference audios and style embedding constraints rather than using only the target audio. Multiple reference audios are automatically selected using the sentence similarity determined by Bidirectional Encoder Representations from Transformers (BERT). In addition, we use ''target'' style embedding from a Pre-trained encoder as a constraint by considering the mutual information between the predicted and ''target'' style embedding. The experimental results show that the proposed model can improve the speech naturalness and content quality with multiple reference audios and can also outperform the baseline model in ABX preference tests of style similarity.

cs.SD↗