Searcharxiv⌕ Search

arXiv subjects

Chen Liu

Publications and source records attributed to Chen Liu.

At least 217 records · Page 12Linked to original sources
arXiv2022

Time-coded Spiking Fourier Transform in Neuromorphic Hardware

After several decades of continuously optimizing computing systems, the Moore's law is reaching itsend. However, there is an increasing demand for fast and efficient processing systems that can handlelarge streams of data while decreasing system footprints. Neuromorphic computing answers thisneed by creating decentralized architectures that communicate with binary events over time. Despiteits rapid growth in the last few years, novel algorithms are needed that can leverage the potential ofthis emerging computing paradigm and can stimulate the design of advanced neuromorphic chips.In this work, we propose a time-based spiking neural network that is mathematically equivalent tothe Fourier transform. We implemented the network in the neuromorphic chip Loihi and conductedexperiments on five different real scenarios with an automotive frequency modulated continuouswave radar. Experimental results validate the algorithm, and we hope they prompt the design of adhoc neuromorphic chips that can improve the efficiency of state-of-the-art digital signal processorsand encourage research on neuromorphic computing for signal processing.

cs.NE↗
arXiv2022

Neural Inertial Localization

This paper proposes the inertial localization problem, the task of estimating the absolute location from a sequence of inertial sensor measurements. This is an exciting and unexplored area of indoor localization research, where we present a rich dataset with 53 hours of inertial sensor data and the associated ground truth locations. We developed a solution, dubbed neural inertial localization (NILoc) which 1) uses a neural inertial navigation technique to turn inertial sensor history to a sequence of velocity vectors; then 2) employs a transformer-based neural architecture to find the device location from the sequence of velocities. We only use an IMU sensor, which is energy efficient and privacy preserving compared to WiFi, cameras, and other data sources. Our approach is significantly faster and achieves competitive results even compared with state-of-the-art methods that require a floorplan and run 20 to 30 times slower. We share our code, model and data at https://sachini.github.io/niloc.

cs.RO↗
arXiv2022

The Fate of Shear-Oscillated Amorphous Solids

The behavior of shear-oscillated amorphous materials is studied using a coarse-grained model. Samples are prepared at different degrees of annealing and then subject to athermal and quasistatic oscillatory deformations at various fixed amplitudes. The steady-state reached after several oscillations is fully determined by the initial preparation and the oscillation amplitude, as seen from stroboscopic stress and energy measurements. Under small oscillations, poorly annealed materials display shear-annealing, while ultra-stabilized materials are insensitive to them. Yet, beyond a critical oscillation amplitude, both kind of materials display a discontinuous transition to the same mixed state composed by a fluid shear-band embedded in a marginal solid. Quantitative relations between uniform shear and the steady-state reached with this protocol are established. The transient regime characterizing the growth and the motion of the shear band is also studied.

cond-mat.soft↗
arXiv2021

Facet Dependent Topological Phase Transition in Bi4Br4

The realization of the coexistence of various topologically nontrivial surface states in one material is expected to lay a foundation for new electric applications with selective robust spin current. Here we apply the magnetoconductivity characteristic and angle-resolved photoemission spectroscopy (ARPES) to visualize the surface-selected electronic features evolution of quasi-one-dimensional material Bi4Br4. The transport measurements indicate the quantum interference correction to conductivity possesses symbolic spin rotational characteristic correlated to the value of Berry phase with the effects of weak localization and weak antilocalization for (001) and (100) surfaces, respectively. The ARPES spectra provide the experimental evidence for quasi-one-dimensional massless Dirac surface state at the side (100) surface and anisotropic massive Dirac surface state at the top (001) surface, respectively, which is highly coincide with the angle-dependent scaling behavior of magnetoconductivity. Our results reveal the facet dependent topological phases in quasi-one-dimensional Bi4Br4, stimulating the further investigations of this dual topology classes and the applications of the feasible technologies of topological spintronics.

cond-mat.mtrl-sci↗
arXiv2021

Multi-objective Scheduling of Electric Vehicle Charging/Discharging with Time of Use Tariff

The increased uptake of electric vehicles (EVs) leads to increased demand for electricity, and sometimes pressure on power grids. Uncoordinated charging of EVs may result in stress on distribution networks, and often some form of optimization is required in the charging process. Optimal coordinated charging is a multi-objective optimization problem (MOOP) in nature, with objective functions such as minimum price charging and minimum disruptions to the grid. In this manuscript, we propose a general multi-objective EV charging/discharging schedule (MOEVCS) framework, where the time of use (TOU) tariff is designed according to the load request at each time stamp. To obtain the optimal scheduling scheme and balance the competing benefits from different stakeholders, such as EV owners, EV charging stations (EVCS), and the grid operator, we design three competing objective functions including EV owner cost, EVCS profit, and the network impact. Moreover, we create four application scenarios with different charging request distributions over the investigated periods. Due to different types of decision variables in this MOOP, we develop a constraint mixed-variable multi-objective evolutionary algorithm (MVMOEA) to implement the proposed MOEVCS framework. Our results demonstrate the effectiveness of MOEVCS in making a balance between three competing objectives.

eess.SY↗
arXiv2021

Large-Gap Quantum Spin Hall State and Temperature-Induced Lifshitz Transition in Bi4Br4

Searching for new quantum spin Hall insulators with large fully opened energy gap to overcome the thermal disturbance at room temperature has attracted tremendous attention due to the one-dimensional (1D) spin-momentum locked topological edge states serving as dissipationless channels for the practical applications in low consumption electronics and high performance spintronics. Here, we report the investigation of topological nature of monolayer Bi4Br4 by the techniques of scanning tunneling microscopy and angle-resolved photoemission spectroscopy (ARPES). The topological non-triviality of 1D edge state integrals within the large bulk energy gap (~ 0.2 eV) is revealed by the first-principle calculations. The ARPES measurements at different temperature show a temperature-induced Lifshitz transition, corresponding to the resistivity anomaly caused by the shift of chemical potential. The connection between the emergency of superconductivity and the Lifshitz transition is discussed.

cond-mat.mtrl-sci↗
arXiv2021

The sub-millimetre non-uniformity measurement of residual and coil-generated field in the magnetic shield using atomic vapor cell

Magnetic field source localization and imaging happen at different scales. The sensing baseline ranges from meter scale such as magnetic anomaly detection, centimeter scale such as brain field imaging to nanometer scale such as the imaging of magnetic skyrmion and single cell. Here we show how atomic vapor cell can be used to realize a baseline of 109.6 μm with a magnetic sensitivity of 10pT/sqrt(Hz)@0.6-100Hz and a dynamic range of 2062-4124nT.We use free induction decay (FID) scheme to suppress low-frequency noise and avoid scale factor variation for different domains due to light non-uniformity. The measurement domains are scanned by digital micro-mirror device (DMD). The currents of 22mA, 30mA, 38mA and 44mA are applied in the coils to generate different fields along the pumping axis which are measured respectively by fitting the FID signals of the probe light. The residual fields of every domain are obtained from the intercept of linearly-fitting of the measurement data corresponding to these four currents. The coil-generated fields are calculated by deducting the residual fields from the total fields. The results demonstrate that the hole of shield affects both the residual and the coil-generated field distribution. The potential impact of field distribution measurement with an outstanding comprehensive properties of spatial resolution, sensitivity and dynamic range is far-reaching. It could lead to capability of 3D magnetography for small stuffs and/or organs in millimeter or even smaller scale.

physics.atom-ph↗
arXiv2021

Training Provably Robust Models by Polyhedral Envelope Regularization

Training certifiable neural networks enables one to obtain models with robustness guarantees against adversarial attacks. In this work, we introduce a framework to bound the adversary-free region in the neighborhood of the input data by a polyhedral envelope, which yields finer-grained certified robustness. We further introduce polyhedral envelope regularization (PER) to encourage larger polyhedral envelopes and thus improve the provable robustness of the models. We demonstrate the flexibility and effectiveness of our framework on standard benchmarks; it applies to networks of different architectures and general activation functions. Compared with the state-of-the-art methods, PER has very little computational overhead and better robustness guarantees without over-regularizing the model.

cs.LG↗
arXiv2021

Hierarchy-Aware T5 with Path-Adaptive Mask Mechanism for Hierarchical Text Classification

Hierarchical Text Classification (HTC), which aims to predict text labels organized in hierarchical space, is a significant task lacking in investigation in natural language processing. Existing methods usually encode the entire hierarchical structure and fail to construct a robust label-dependent model, making it hard to make accurate predictions on sparse lower-level labels and achieving low Macro-F1. In this paper, we propose a novel PAMM-HiA-T5 model for HTC: a hierarchy-aware T5 model with path-adaptive mask mechanism that not only builds the knowledge of upper-level labels into low-level ones but also introduces path dependency information in label prediction. Specifically, we generate a multi-level sequential label structure to exploit hierarchical dependency across different levels with Breadth-First Search (BFS) and T5 model. To further improve label dependency prediction within each path, we then propose an original path-adaptive mask mechanism (PAMM) to identify the label's path information, eliminating sources of noises from other paths. Comprehensive experiments on three benchmark datasets show that our novel PAMM-HiA-T5 model greatly outperforms all state-of-the-art HTC approaches especially in Macro-F1. The ablation studies show that the improvements mainly come from our innovative approach instead of T5.

cs.CL↗
arXiv2021

FLiText: A Faster and Lighter Semi-Supervised Text Classification with Convolution Networks

In natural language processing (NLP), state-of-the-art (SOTA) semi-supervised learning (SSL) frameworks have shown great performance on deep pre-trained language models such as BERT, and are expected to significantly reduce the demand for manual labeling. However, our empirical studies indicate that these frameworks are not suitable for lightweight models such as TextCNN, LSTM and etc. In this work, we develop a new SSL framework called FLiText, which stands for Faster and Lighter semi-supervised Text classification. FLiText introduces an inspirer network together with the consistency regularization framework, which leverages a generalized regular constraint on the lightweight models for efficient SSL. As a result, FLiText obtains new SOTA performance for lightweight models across multiple SSL benchmarks on text classification. Compared with existing SOTA SSL methods on TextCNN, FLiText improves the accuracy of lightweight model TextCNN from 51.00% to 90.49% on IMDb, 39.8% to 58.06% on Yelp-5, and from 55.3% to 65.08% on Yahoo. In addition, compared with the fully supervised method on the full dataset, FLiText just uses less than 1% of labeled data to improve the accuracy by 6.59%, 3.94%, and 3.22% on the datasets of IMDb, Yelp-5, and Yahoo respectively.

cs.CL↗
arXiv2021

Cross-Site Severity Assessment of COVID-19 from CT Images via Domain Adaptation

Early and accurate severity assessment of Coronavirus disease 2019 (COVID-19) based on computed tomography (CT) images offers a great help to the estimation of intensive care unit event and the clinical decision of treatment planning. To augment the labeled data and improve the generalization ability of the classification model, it is necessary to aggregate data from multiple sites. This task faces several challenges including class imbalance between mild and severe infections, domain distribution discrepancy between sites, and presence of heterogeneous features. In this paper, we propose a novel domain adaptation (DA) method with two components to address these problems. The first component is a stochastic class-balanced boosting sampling strategy that overcomes the imbalanced learning problem and improves the classification performance on poorly-predicted classes. The second component is a representation learning that guarantees three properties: 1) domain-transferability by prototype triplet loss, 2) discriminant by conditional maximum mean discrepancy loss, and 3) completeness by multi-view reconstruction loss. Particularly, we propose a domain translator and align the heterogeneous data to the estimated class prototypes (i.e., class centers) in a hyper-sphere manifold. Experiments on cross-site severity assessment of COVID-19 from CT images show that the proposed method can effectively tackle the imbalanced learning problem and outperform recent DA approaches.

eess.IV↗
arXiv2021

On the Dynamics of Liquids in the Large-Dimensional Limit

In this work, we analytically derive the exact closed dynamical equations for a liquid with short-ranged interactions in large spatial dimensions using the same statistical mechanics tools employed to analyze Brownian motion. Our derivation greatly simplifies the original path-integral-based route to these equations and provides new insight into the physical features associated with high-dimensional liquids and glass formation. Most importantly, our construction provides a facile route to the exact dynamical analysis of important related dynamical problems, as well as a means to devise cluster generalizations of the exact solution in infinite dimensions. This latter fact opens the door to the construction of increasingly accurate theories of vitrification in three-dimensional liquids.

cond-mat.stat-mech↗
arXiv2021

Fourier Transform Approximation as an Auxiliary Task for Image Classification

Image reconstruction is likely the most predominant auxiliary task for image classification, but we would like to think twice about this convention. In this paper, we investigated "approximating the Fourier Transform of the input image" as a potential alternative, in the hope that it may further boost the performances on the primary task or introduce novel constraints not well covered by image reconstruction. We experimented with five popular classification architectures on the CIFAR-10 dataset, and the empirical results indicated that our proposed auxiliary task generally improves the classification accuracy. More notably, the results showed that in certain cases our proposed auxiliary task may enhance the classifiers' resistance to adversarial attacks generated using the fast gradient sign method.

cs.CV↗
arXiv2021

MG-DVD: A Real-time Framework for Malware Variant Detection Based on Dynamic Heterogeneous Graph Learning

Detecting the newly emerging malware variants in real time is crucial for mitigating cyber risks and proactively blocking intrusions. In this paper, we propose MG-DVD, a novel detection framework based on dynamic heterogeneous graph learning, to detect malware variants in real time. Particularly, MG-DVD first models the fine-grained execution event streams of malware variants into dynamic heterogeneous graphs and investigates real-world meta-graphs between malware objects, which can effectively characterize more discriminative malicious evolutionary patterns between malware and their variants. Then, MG-DVD presents two dynamic walk-based heterogeneous graph learning methods to learn more comprehensive representations of malware variants, which significantly reduces the cost of the entire graph retraining. As a result, MG-DVD is equipped with the ability to detect malware variants in real time, and it presents better interpretability by introducing meaningful meta-graphs. Comprehensive experiments on large-scale samples prove that our proposed MG-DVD outperforms state-of-the-art methods in detecting malware variants in terms of effectiveness and efficiency.

cs.CR↗
arXiv2021

Deep Learning Identifies Neuroimaging Signatures of Alzheimer's Disease Using Structural and Synthesized Functional MRI Data

Current neuroimaging techniques provide paths to investigate the structure and function of the brain in vivo and have made great advances in understanding Alzheimer's disease (AD). However, the group-level analyses prevalently used for investigation and understanding of the disease are not applicable for diagnosis of individuals. More recently, deep learning, which can efficiently analyze large-scale complex patterns in 3D brain images, has helped pave the way for computer-aided individual diagnosis by providing accurate and automated disease classification. Great progress has been made in classifying AD with deep learning models developed upon increasingly available structural MRI data. The lack of scale-matched functional neuroimaging data prevents such models from being further improved by observing functional changes in pathophysiology. Here we propose a potential solution by first learning a structural-to-functional transformation in brain MRI, and further synthesizing spatially matched functional images from large-scale structural scans. We evaluated our approach by building computational models to discriminate patients with AD from healthy normal subjects and demonstrated a performance boost after combining the structural and synthesized functional brain images into the same model. Furthermore, our regional analyses identified the temporal lobe to be the most predictive structural-region and the parieto-occipital lobe to be the most predictive functional-region of our model, which are both in concordance with previous group-level neuroimaging findings. Together, we demonstrate the potential of deep learning with large-scale structural and synthesized functional MRI to impact AD classification and to identify AD's neuroimaging signatures.

q-bio.QM↗
arXiv2021

Learning Dynamic Alignment via Meta-filter for Few-shot Learning

Few-shot learning (FSL), which aims to recognise new classes by adapting the learned knowledge with extremely limited few-shot (support) examples, remains an important open problem in computer vision. Most of the existing methods for feature alignment in few-shot learning only consider image-level or spatial-level alignment while omitting the channel disparity. Our insight is that these methods would lead to poor adaptation with redundant matching, and leveraging channel-wise adjustment is the key to well adapting the learned knowledge to new classes. Therefore, in this paper, we propose to learn a dynamic alignment, which can effectively highlight both query regions and channels according to different local support information. Specifically, this is achieved by first dynamically sampling the neighbourhood of the feature position conditioned on the input few shot, based on which we further predict a both position-dependent and channel-dependent Dynamic Meta-filter. The filter is used to align the query feature with position-specific and channel-specific knowledge. Moreover, we adopt Neural Ordinary Differential Equation (ODE) to enable a more accurate control of the alignment. In such a sense our model is able to better capture fine-grained semantic context of the few-shot example and thus facilitates dynamical knowledge adaptation for few-shot learning. The resulting framework establishes the new state-of-the-arts on major few-shot visual recognition benchmarks, including miniImageNet and tieredImageNet.

cs.CV↗