arXiv subjects

Qian Chen

Publications and source records attributed to Qian Chen.

At least 307 records · Page 17Linked to original sources

arXiv2022

Spin network entanglement: coarse-graining

In loop quantum gravity, partitioning graph introduces boundaries and entanglement between spin sub-networks, reflecting non-local degrees of freedom and correlation amongst spatial regions. This gives rise to the view of coarse-graining, reducing the degrees of freedom that are unnecessary to be considered. The present work sets coarse-graining in the framework of bulk-boundary relation. We investigates the spin network entanglement, showing that the entanglement is invariant under the coarse-graining at kinematical level. Moreover, we build the transformation between holonomy operators based on finer graph and coarser graph, and reveal the preservation of the coarse-graining method under the evolution generated by the holonomy operator. This leads to a holographical perspective for the entanglement issue in loop quantum gravity.

hep-th↗

arXiv2022

Critical phenomena in dynamical scalarization of charged black hole

We report a new black hole scalarization mechanism and disclose novel dynamical critical phenomena in the process of the nonlinear accretion of the scalar field into black holes. The accretion process can transform a seed black hole into a final scalarized or bald black hole, depending on the initial parameter of the scalar field $p$. There is a critical parameter $p_{\ast}$ and near it all intermediate solutions are attracted to a critical solution and stay there for a time scaling as $T\propto-γ\ln|p-p_{\ast}|$. At late times, the solutions evolve into scalarized black holes if $p>p_{\ast}$, or bald black holes if $p<p_{\ast}$. The final masses of the resulting scalarized/bald black holes satisfy power-laws $M_{p}-M_{\pm}\propto|p-p_{\ast}|^{γ_{\pm}}$ where $M_{\pm}$ are the masses of the scalarized/bald black holes when $p\to p_\ast$ from above/below, and $γ_{\pm}$ the corresponding exponents.

gr-qc↗

arXiv2022

Intertwiner Entanglement Excitation and Holonomy Operator

In the loop quantum gravity framework, spin network states carry entanglement between quantum excitations of the geometry at different space points. This intertwiner entanglement is gauge-invariant and comes from quantum superposition of spins and intertwiners. Bipartite entanglement can be interpreted as a witness of distance, while multipartite entanglement reflects the curvature of the quantum geometry. The present work investigates how the bipartite and multipartite intertwiner entanglement changes under the action of the holonomy operator, which is the basic building block of loop quantum gravity's dynamics. We reveal the relation between entanglement excitation and the dispersion of the holonomy operator. This leads to a new interesting connection between bulk geometry and boundary observables via the dynamics of entanglement.

gr-qc↗

arXiv2022

Weakly Supervised Object Localization as Domain Adaption

Weakly supervised object localization (WSOL) focuses on localizing objects only with the supervision of image-level classification masks. Most previous WSOL methods follow the classification activation map (CAM) that localizes objects based on the classification structure with the multi-instance learning (MIL) mechanism. However, the MIL mechanism makes CAM only activate discriminative object parts rather than the whole object, weakening its performance for localizing objects. To avoid this problem, this work provides a novel perspective that models WSOL as a domain adaption (DA) task, where the score estimator trained on the source/image domain is tested on the target/pixel domain to locate objects. Under this perspective, a DA-WSOL pipeline is designed to better engage DA approaches into WSOL to enhance localization performance. It utilizes a proposed target sampling strategy to select different types of target samples. Based on these types of target samples, domain adaption localization (DAL) loss is elaborated. It aligns the feature distribution between the two domains by DA and makes the estimator perceive target domain cues by Universum regularization. Experiments show that our pipeline outperforms SOTA methods on multi benchmarks. Code are released at \url{https://github.com/zh460045050/DA-WSOL_CVPR2022}.

cs.CV↗

arXiv2022

FinRL: A Deep Reinforcement Learning Library for Automated Stock Trading in Quantitative Finance

As deep reinforcement learning (DRL) has been recognized as an effective approach in quantitative finance, getting hands-on experiences is attractive to beginners. However, to train a practical DRL trading agent that decides where to trade, at what price, and what quantity involves error-prone and arduous development and debugging. In this paper, we introduce a DRL library FinRL that facilitates beginners to expose themselves to quantitative finance and to develop their own stock trading strategies. Along with easily-reproducible tutorials, FinRL library allows users to streamline their own developments and to compare with existing schemes easily. Within FinRL, virtual environments are configured with stock market datasets, trading agents are trained with neural networks, and extensive backtesting is analyzed via trading performance. Moreover, it incorporates important trading constraints such as transaction cost, market liquidity and the investor's degree of risk-aversion. FinRL is featured with completeness, hands-on tutorial and reproducibility that favors beginners: (i) at multiple levels of time granularity, FinRL simulates trading environments across various stock markets, including NASDAQ-100, DJIA, S&P 500, HSI, SSE 50, and CSI 300; (ii) organized in a layered architecture with modular structure, FinRL provides fine-tuned state-of-the-art DRL algorithms (DQN, DDPG, PPO, SAC, A2C, TD3, etc.), commonly-used reward functions and standard evaluation baselines to alleviate the debugging workloads and promote the reproducibility, and (iii) being highly extendable, FinRL reserves a complete set of user-import interfaces. Furthermore, we incorporated three application demonstrations, namely single stock trading, multiple stock trading, and portfolio allocation. The FinRL library will be available on Github at link https://github.com/AI4Finance-LLC/FinRL-Library.

q-fin.TR↗

arXiv2022

ProsoSpeech: Enhancing Prosody With Quantized Vector Pre-training in Text-to-Speech

Expressive text-to-speech (TTS) has become a hot research topic recently, mainly focusing on modeling prosody in speech. Prosody modeling has several challenges: 1) the extracted pitch used in previous prosody modeling works have inevitable errors, which hurts the prosody modeling; 2) different attributes of prosody (e.g., pitch, duration and energy) are dependent on each other and produce the natural prosody together; and 3) due to high variability of prosody and the limited amount of high-quality data for TTS training, the distribution of prosody cannot be fully shaped. To tackle these issues, we propose ProsoSpeech, which enhances the prosody using quantized latent vectors pre-trained on large-scale unpaired and low-quality text and speech data. Specifically, we first introduce a word-level prosody encoder, which quantizes the low-frequency band of the speech and compresses prosody attributes in the latent prosody vector (LPV). Then we introduce an LPV predictor, which predicts LPV given word sequence. We pre-train the LPV predictor on large-scale text and low-quality speech data and fine-tune it on the high-quality TTS dataset. Finally, our model can generate expressive speech conditioned on the predicted LPV. Experimental results show that ProsoSpeech can generate speech with richer prosody compared with baseline methods.

eess.AS↗

arXiv2021

Background-aware Classification Activation Map for Weakly Supervised Object Localization

Weakly supervised object localization (WSOL) relaxes the requirement of dense annotations for object localization by using image-level classification masks to supervise its learning process. However, current WSOL methods suffer from excessive activation of background locations and need post-processing to obtain the localization mask. This paper attributes these issues to the unawareness of background cues, and propose the background-aware classification activation map (B-CAM) to simultaneously learn localization scores of both object and background with only image-level labels. In our B-CAM, two image-level features, aggregated by pixel-level features of potential background and object locations, are used to purify the object feature from the object-related background and to represent the feature of the pure-background sample, respectively. Then based on these two features, both the object classifier and the background classifier are learned to determine the binary object localization mask. Our B-CAM can be trained in end-to-end manner based on a proposed stagger classification loss, which not only improves the objects localization but also suppresses the background activation. Experiments show that our B-CAM outperforms one-stage WSOL methods on the CUB-200, OpenImages and VOC2012 datasets.

cs.CV↗

arXiv2021

Super-resolution reconstruction of cytoskeleton image based on A-net deep learning network

To date, live-cell imaging at the nanometer scale remains challenging. Even though super-resolution microscopy methods have enabled visualization of subcellular structures below the optical resolution limit, the spatial resolution is still far from enough for the structural reconstruction of biomolecules in vivo (i.e. ~24 nm thickness of microtubule fiber). In this study, we proposed an A-net network and showed that the resolution of cytoskeleton images captured by a confocal microscope can be significantly improved by combining the A-net deep learning network with the DWDC algorithm based on degradation model. Utilizing the DWDC algorithm to construct new datasets and taking advantage of A-net neural network's features (i.e., considerably fewer layers), we successfully removed the noise and flocculent structures, which originally interfere with the cellular structure in the raw image, and improved the spatial resolution by 10 times using relatively small dataset. We, therefore, conclude that the proposed algorithm that combines A-net neural network with the DWDC method is a suitable and universal approach for exacting structural details of biomolecules, cells and organs from low-resolution images.

eess.IV↗

arXiv2021

TRS: Transferability Reduced Ensemble via Encouraging Gradient Diversity and Model Smoothness

Adversarial Transferability is an intriguing property - adversarial perturbation crafted against one model is also effective against another model, while these models are from different model families or training processes. To better protect ML systems against adversarial attacks, several questions are raised: what are the sufficient conditions for adversarial transferability and how to bound it? Is there a way to reduce the adversarial transferability in order to improve the robustness of an ensemble ML model? To answer these questions, in this work we first theoretically analyze and outline sufficient conditions for adversarial transferability between models; then propose a practical algorithm to reduce the transferability between base models within an ensemble to improve its robustness. Our theoretical analysis shows that only promoting the orthogonality between gradients of base models is not enough to ensure low transferability; in the meantime, the model smoothness is an important factor to control the transferability. We also provide the lower and upper bounds of adversarial transferability under certain conditions. Inspired by our theoretical analysis, we propose an effective Transferability Reduced Smooth(TRS) ensemble training strategy to train a robust ensemble with low transferability by enforcing both gradient orthogonality and model smoothness between base models. We conduct extensive experiments on TRS and compare with 6 state-of-the-art ensemble baselines against 8 whitebox attacks on different datasets, demonstrating that the proposed TRS outperforms all baselines significantly.

cs.LG↗

arXiv2021

Learning Synergistic Attention for Light Field Salient Object Detection

We propose a novel Synergistic Attention Network (SA-Net) to address the light field salient object detection by establishing a synergistic effect between multi-modal features with advanced attention mechanisms. Our SA-Net exploits the rich information of focal stacks via 3D convolutional neural networks, decodes the high-level features of multi-modal light field data with two cascaded synergistic attention modules, and predicts the saliency map using an effective feature fusion module in a progressive manner. Extensive experiments on three widely-used benchmark datasets show that our SA-Net outperforms 28 state-of-the-art models, sufficiently demonstrating its effectiveness and superiority. Our code is available at https://github.com/PanoAsh/SA-Net.

cs.CV↗

arXiv2021

Sequence Model with Self-Adaptive Sliding Window for Efficient Spoken Document Segmentation

Transcripts generated by automatic speech recognition (ASR) systems for spoken documents lack structural annotations such as paragraphs, significantly reducing their readability. Automatically predicting paragraph segmentation for spoken documents may both improve readability and downstream NLP performance such as summarization and machine reading comprehension. We propose a sequence model with self-adaptive sliding window for accurate and efficient paragraph segmentation. We also propose an approach to exploit phonetic information, which significantly improves robustness of spoken document segmentation to ASR errors. Evaluations are conducted on the English Wiki-727K document segmentation benchmark, a Chinese Wikipedia-based document segmentation dataset we created, and an in-house Chinese spoken document dataset. Our proposed model outperforms the state-of-the-art (SOTA) model based on the same BERT-Base, increasing segmentation F1 on the English benchmark by 4.2 points and on Chinese datasets by 4.3-10.1 points, while reducing inference time to less than 1/6 of inference time of the current SOTA.

cs.CL↗

arXiv2021

Neighbor Discovery for VANET with Gossip Mechanism and Multi-packet Reception

Neighbor discovery (ND) is a key initial step of network configuration and prerequisite of vehicular ad hoc network (VANET). However, the convergence efficiency of ND is facing the requirements of multi-vehicle fast networking of VANET with frequent topology changes. This paper proposes gossip-based information dissemination and sensing information assisted ND with MPR (GSIM-ND) algorithm for VANET. GSIM-ND algorithm leverages efficient gossip-based information dissemination in the case of multi-packet reception (MPR). Besides, through multi-target detection function of multiple sensors installed in roadside unit (RSU), RSU can sense the distribution of vehicles and help vehicles to obtain the distribution of their neighbors. Thus, GSIM-ND algorithm leverages the dissemination of sensing information as well. The expected number of discovered neighbors within a given period is theoretically derived and used as the critical metric to evaluate the performance of GSIM-ND algorithm. The expected bounds of the number of time slots when a given number of neighbors needs to be discovered is derived as well. The simulation results verify the correctness of theoretical derivation. It is discovered that GSIM-ND algorithm proposed in this paper can always reach the short-term convergence quickly. Moreover, GSIM-ND algorithm is more efficient and stable compared with completely random algorithm (CRA), scan-based algorithm (SBA) and gossip-based algorithm. The convergence time of GSIM-ND algorithm is 40\%-90\% lower than that of these existing algorithms for both low density and high density networks. Thus, GSIM-ND can improve the efficiency of ND algorithm.

cs.NI↗

arXiv2021

BeamTransformer: Microphone Array-based Overlapping Speech Detection

We propose BeamTransformer, an efficient architecture to leverage beamformer's edge in spatial filtering and transformer's capability in context sequence modeling. BeamTransformer seeks to optimize modeling of sequential relationship among signals from different spatial direction. Overlapping speech detection is one of the tasks where such optimization is favorable. In this paper we effectively apply BeamTransformer to detect overlapping segments. Comparing to single-channel approach, BeamTransformer exceeds in learning to identify the relationship among different beam sequences and hence able to make predictions not only from the acoustic signals but also the localization of the source. The results indicate that a successful incorporation of microphone array signals can lead to remarkable gains. Moreover, BeamTransformer takes one step further, as speech from overlapped speakers have been internally separated into different beams.

cs.SD↗

arXiv2021

Pre-training for Spoken Language Understanding with Joint Textual and Phonetic Representation Learning

In the traditional cascading architecture for spoken language understanding (SLU), it has been observed that automatic speech recognition errors could be detrimental to the performance of natural language understanding. End-to-end (E2E) SLU models have been proposed to directly map speech input to desired semantic frame with a single model, hence mitigating ASR error propagation. Recently, pre-training technologies have been explored for these E2E models. In this paper, we propose a novel joint textual-phonetic pre-training approach for learning spoken language representations, aiming at exploring the full potentials of phonetic information to improve SLU robustness to ASR errors. We explore phoneme labels as high-level speech features, and design and compare pre-training tasks based on conditional masked language model objectives and inter-sentence relation objectives. We also investigate the efficacy of combining textual and phonetic information during fine-tuning. Experimental results on spoken language understanding benchmarks, Fluent Speech Commands and SNIPS, show that the proposed approach significantly outperforms strong baseline models and improves robustness of spoken language understanding to ASR errors.

cs.CL↗

arXiv2021

PPT: A Privacy-Preserving Global Model Training Protocol for Federated Learning in P2P Networks

The concept of Federated Learning (FL) has emerged as a convergence of machine learning, information, and communication technology. It is vital to the development of machine learning, which is expected to be fully decentralized, privacy-preserving, secure, and robust. However, general federated learning settings with a central server can't meet requirements in decentralized environment. In this paper, we propose a decentralized, secure and privacy-preserving global model training protocol, named PPT, for federated learning in Peer-to-peer (P2P) Networks. PPT uses a one-hop communication form to aggregate local model update parameters and adopts the symmetric cryptosystem to ensure security. It is worth mentioning that PPT modifies the Eschenauer-Gligor (E-G) scheme to distribute keys for encryption. In terms of privacy preservation, PPT generates random noise to disturb local model update parameters. The noise is eliminated ultimately, which ensures the global model performance compared with other noise-based privacy-preserving methods in FL, e.g., differential privacy. PPT also adopts Game Theory to resist collusion attacks. Through extensive analysis, we demonstrate that PPT various security threats and preserve user privacy. Ingenious experiments demonstrate the utility and efficiency as well.

cs.CR↗

arXiv2021

Towards Making Deep Learning-based Vulnerability Detectors Robust

Automatically detecting software vulnerabilities in source code is an important problem that has attracted much attention. In particular, deep learning-based vulnerability detectors, or DL-based detectors, are attractive because they do not need human experts to define features or patterns of vulnerabilities. However, such detectors' robustness is unclear. In this paper, we initiate the study in this aspect by demonstrating that DL-based detectors are not robust against simple code transformations, dubbed attacks in this paper, as these transformations may be leveraged for malicious purposes. As a first step towards making DL-based detectors robust against such attacks, we propose an innovative framework, dubbed ZigZag, which is centered at (i) decoupling feature learning and classifier learning and (ii) using a ZigZag-style strategy to iteratively refine them until they converge to robust features and robust classifiers. Experimental results show that the ZigZag framework can substantially improve the robustness of DL-based detectors.

cs.CR↗

arXiv2021

From Single to Multiple: Leveraging Multi-level Prediction Spaces for Video Forecasting

Despite video forecasting has been a widely explored topic in recent years, the mainstream of the existing work still limits their models with a single prediction space but completely neglects the way to leverage their model with multi-prediction spaces. This work fills this gap. For the first time, we deeply study numerous strategies to perform video forecasting in multi-prediction spaces and fuse their results together to boost performance. The prediction in the pixel space usually lacks the ability to preserve the semantic and structure content of the video however the prediction in the high-level feature space is prone to generate errors in the reduction and recovering process. Therefore, we build a recurrent connection between different feature spaces and incorporate their generations in the upsampling process. Rather surprisingly, this simple idea yields a much more significant performance boost than PhyDNet (performance improved by 32.1% MAE on MNIST-2 dataset, and 21.4% MAE on KTH dataset). Both qualitative and quantitative evaluations on four datasets demonstrate the generalization ability and effectiveness of our approach. We show that our model significantly reduces the troublesome distortions and blurry artifacts and brings remarkable improvements to the accuracy in long term video prediction. The code will be released soon.

cs.CV↗

arXiv2021

Chiral emergence in multistep hierarchical assembly of achiral conjugated polymers

Intimately connected to the rule of life, chirality remains a long-time fascination in biology, chemistry, physics and materials science. Chiral structures, e.g., nucleic acid and cholesteric phase developed from chiral molecules are common in nature and synthetic soft materials. While it was recently discovered that achiral but bent core mesogens can also form chiral helices, the assembly of chiral microstructures from achiral polymers has rarely been explored. Here, we reveal chiral emergence from achiral conjugated polymers for the first time, in which hierarchical helical structures are developed through a multistep assembly pathway. Upon increasing concentration beyond a threshold volume fraction, pre-aggregated polymer nanofibers form lyotropic liquid crystalline (LC) mesophases with complex, chiral morphologies. Combining imaging, X-ray and spectroscopy techniques with molecular simulations, we demonstrate that this structural evolution arises from torsional polymer molecules which induce multiscale helical assembly, progressing from nano- to micron scale helical structures as the solution concentration increases. This study unveils a previously unknown complex state of matter for conjugated polymers that can pave way to a new field of chiral (opto)electronics. We anticipate that hierarchical chiral helical structures can profoundly impact how conjugated polymers interact with light, transport charges, and transduce signals from biomolecular interactions and even give rise to properties unimagined before.

cond-mat.soft↗