Search arXiv⌕ Search

arXiv subjects

Yan Liu

Publications and source records attributed to Yan Liu.

At least 307 records · Page 17Linked to original sources
arXiv2022

3D-Morphomics, Morphological Features on CT scans for lung nodule malignancy diagnosis

Pathologies systematically induce morphological changes, thus providing a major but yet insufficiently quantified source of observables for diagnosis. The study develops a predictive model of the pathological states based on morphological features (3D-morphomics) on Computed Tomography (CT) volumes. A complete workflow for mesh extraction and simplification of an organ's surface is developed, and coupled with an automatic extraction of morphological features given by the distribution of mean curvature and mesh energy. An XGBoost supervised classifier is then trained and tested on the 3D-morphomics to predict the pathological states. This framework is applied to the prediction of the malignancy of lung's nodules. On a subset of NLST database with malignancy confirmed biopsy, using 3D-morphomics only, the classification model of lung nodules into malignant vs. benign achieves 0.964 of AUC. Three other sets of classical features are trained and tested, (1) clinical relevant features gives an AUC of 0.58, (2) 111 radiomics gives an AUC of 0.976, (3) radiologist ground truth (GT) containing the nodule size, attenuation and spiculation qualitative annotations gives an AUC of 0.979. We also test the Brock model and obtain an AUC of 0.826. Combining 3D-morphomics and radiomics features achieves state-of-the-art results with an AUC of 0.978 where the 3D-morphomics have some of the highest predictive powers. As a validation on a public independent cohort, models are applied to the LIDC dataset, the 3D-morphomics achieves an AUC of 0.906 and the 3D-morphomics+radiomics achieves an AUC of 0.958, which ranks second in the challenge among deep models. It establishes the curvature distributions as efficient features for predicting lung nodule malignancy and a new method that can be applied directly to arbitrary computer aided diagnosis task.

eess.IV↗
arXiv2022

SOUL-Net: A Sparse and Low-Rank Unrolling Network for Spectral CT Image Reconstruction

Spectral computed tomography (CT) is an emerging technology, that generates a multienergy attenuation map for the interior of an object and extends the traditional image volume into a 4D form. Compared with traditional CT based on energy-integrating detectors, spectral CT can make full use of spectral information, resulting in high resolution and providing accurate material quantification. Numerous model-based iterative reconstruction methods have been proposed for spectral CT reconstruction. However, these methods usually suffer from difficulties such as laborious parameter selection and expensive computational costs. In addition, due to the image similarity of different energy bins, spectral CT usually implies a strong low-rank prior, which has been widely adopted in current iterative reconstruction models. Singular value thresholding (SVT) is an effective algorithm to solve the low-rank constrained model. However, the SVT method requires manual selection of thresholds, which may lead to suboptimal results. To relieve these problems, in this paper, we propose a Sparse and lOw-rank UnroLling Network for spectral CT image reconstruction (SOUL-Net), that learns the parameters and thresholds in a data-driven manner. Furthermore, a Taylor expansion-based neural network backpropagation method is introduced to improve the numerical stability. The qualitative and quantitative results demonstrate that the proposed method outperforms several representative state-of-the-art algorithms in terms of detail preservation and artifact reduction.

physics.med-ph↗
arXiv2022

A note on asymptotic profiles for the thermoelastic plate system

We investigate the Cauchy problem for the thermoelastic plate system associated with Newton's law of cooling, where optimal growth ($n\leqslant 4$) or decay ($n\geqslant 5$) estimates and asymptotic profiles of solutions for large-time are studied. Especially, the additional lower-order term in the temperature equation weakens decay rates of the vertical displacement, and leads to a new leading term comparing with the classical thermoelastic plates.

math.AP↗
arXiv2022

Stability limits for modes held in alternating trapping-expulsive potentials

We elaborate a scheme of trapping-expulsion management (TEM), in the form of the quadratic potential periodically switching between confinement and expulsion, as a means of stabilization of two-dimensional dynamical states against the backdrop of the critical collapse driven by the cubic self-attraction with strength g. The TEM scheme may be implemented, as spatially or temporally periodic modulations, in optics or BEC, respectively. The consideration is carried out by dint of numerical simulations and variational approximation (VA). In terms of the VA, the dynamics amounts to a nonlinear Ermakov equation, which, in turn, is tantamount to a linear Mathieu equation. Stability boundaries are found as functions of g and parameters of the periodic modulation of the trapping potential. Below the usual collapse threshold, which is known, in the numerical form, as g < 5.85 (in the standard notation), the stability is limited by the onset of the parametric resonance. This stability limit, including the setup with the self-repulsive sign of the cubic term (g < 0), is accurately predicted by the VA. At g > 5.85, the collapse threshold is found with the help of full numerical simulations. The relative increase of the critical value of g above 5.85 is ~ 1.5%, which is a meaningful result, even if its size is small, because the collapse threshold is a universal constant, which is difficult to change.

cond-mat.quant-gas↗
arXiv2022

Electrically tunable second harmonic generation in atomically thin ReS2

Electrical tuning of second-order nonlinearity in optical materials is attractive to strengthen and expand the functionalities of nonlinear optical technologies, though its implementation remains elusive. Here, we report the electrically tunable second-order nonlinearity in atomically thin ReS2 flakes benefiting from their distorted 1T crystal structure and interlayer charge transfer. Enabled by the efficient electrostatic control of the few-atomic-layer ReS2, we show that second harmonic generation (SHG) can be induced in odd-number-layered ReS2 flakes which are centrosymmetric and thus without intrinsic SHG. Moreover, the SHG can be precisely modulated by the electric field, reversibly switching from almost zero to an amplitude more than one order of magnitude stronger than that of the monolayer MoS2. For the even-number-layered ReS2 flakes with the intrinsic SHG, the external electric field could be leveraged to enhance the SHG. We further perform the first-principles calculations which suggest that the modification of in-plane second-order hyperpolarizability by the redistributed interlayer-transferring charges in the distorted 1T crystal structure underlies the electrically tunable SHG in ReS2. With its active SHG tunability while using the facile electrostatic control, our work may further expand the nonlinear optoelectronic functions of two-dimensional materials for developing electrically controllable nonlinear optoelectronic devices.

physics.optics↗
arXiv2022

A Multi-scale Time-series Dataset with Benchmark for Machine Learning in Decarbonized Energy Grids

The electric grid is a key enabling infrastructure for the ambitious transition towards carbon neutrality as we grapple with climate change. With deepening penetration of renewable energy resources and electrified transportation, the reliable and secure operation of the electric grid becomes increasingly challenging. In this paper, we present PSML, a first-of-its-kind open-access multi-scale time-series dataset, to aid in the development of data-driven machine learning (ML) based approaches towards reliable operation of future electric grids. The dataset is generated through a novel transmission + distribution (T+D) co-simulation designed to capture the increasingly important interactions and uncertainties of the grid dynamics, containing electric load, renewable generation, weather, voltage and current measurements over multiple spatio-temporal scales. Using PSML, we provide state-of-the-art ML baselines on three challenging use cases of critical importance to achieve: (i) early detection, accurate classification and localization of dynamic disturbance events; (ii) robust hierarchical forecasting of load and renewable energy with the presence of uncertainties and extreme events; and (iii) realistic synthetic generation of physical-law-constrained measurement time series. We envision that this dataset will enable advances for ML in dynamic systems, while simultaneously allowing ML researchers to contribute towards carbon-neutral electricity and mobility.

cs.LG↗
arXiv2022

A Search for Low-mass Dark Matter via Bremsstrahlung Radiation and the Migdal Effect in SuperCDMS

In this paper, we present a re-analysis of SuperCDMS data using a profile likelihood approach to search for sub-GeV dark matter particles (DM) through two inelastic scattering channels: bremsstrahlung radiation and the Migdal effect. By considering possible inelastic scattering channels, experimental sensitivity can be extended to DM masses that would otherwise be undetectable through the DM-nucleon elastic scattering channel, given the energy threshold of current experiments. We exclude DM masses down to $220~\textrm{MeV}/c^2$ at $2.7 \times 10^{-30}~\textrm{cm}^2$ via the bremsstrahlung channel. The Migdal channel search excludes DM masses down to $30~\textrm{MeV}/c^2$ at $5.0 \times 10^{-30}~\textrm{cm}^2$.

hep-ex↗
arXiv2022

Forecasting Loss of Signal in Optical Networks with Machine Learning

Loss of Signal (LOS) represents a significant cost for operators of optical networks. By studying large sets of real-world Performance Monitoring (PM) data collected from six international optical networks, we find that it is possible to forecast LOS events with good precision 1-7 days before they occur, albeit at relatively low recall, with supervised machine learning (ML). Our study covers twelve facility types, including 100G lines and ETH10G clients. We show that the precision for a given network improves when training on multiple networks simultaneously relative to training on an individual network. Furthermore, we show that it is possible to forecast LOS from all facility types and all networks with a single model, whereas fine-tuning for a particular facility or network only brings modest improvements. Hence our ML models remain effective for optical networks previously unknown to the model, which makes them usable for commercial applications.

cs.NI↗
arXiv2022

Topological hydrodynamic modes and holography

We study topological modes in relativistic hydrodynamics by weakly breaking the conservation of energy momentum tensor. Several systems have been found to have topologically nontrivial crossing nodes in the spectrum of hydrodynamic modes and some of them are only topologically nontrivial with the protection of reflection symmetries in two directions. The nontrivial topology for all these systems is further confirmed from a calculation of the topological invariant. Associated transport properties and second order effects have also been studied for these systems. The non-conservation terms of the energy momentum tensor could come from an external effective symmetric tensor matter field or a gravitational field which could be generated by a specific non-inertial reference frame transformation from the original inertial reference frame. Finally we introduce a possible holographic realization of one of these systems. We propose a new method to calculate the holographic Ward identities for the energy momentum tensor without calculating out all components of the Green functions and match the Ward identities of both sides.

hep-th↗
arXiv2022

COVID-19 Vaccine Misinformation Campaigns and Social Media Narratives

COVID-19 vaccine hesitancy has increased concerns about vaccine uptake required to overcome the pandemic and protect public health. A critical factor associated with anti-vaccine attitudes is the information shared on social media. In this work, we investigate misinformation communities and narratives that can contribute to COVID-19 vaccine hesitancy. During the pandemic, anti-science and political misinformation/conspiracies have been rampant on social media. Therefore, we investigate misinformation and conspiracy groups and their characteristic behaviours in Twitter data collected on COVID-19 vaccines. We identify if any suspicious coordinated efforts are present in promoting vaccine misinformation, and find two suspicious groups - one promoting a 'Great Reset' conspiracy which suggests that the pandemic is orchestrated by world leaders to take control of the economy, with vaccine related misinformation and strong anti-vaccine and anti-social messages such as no lock-downs; and another promoting the Bioweapon theory. Misinformation promoted is largely from the anti-vaccine and far-right communities in the 3-core of the retweet graph, with its tweets proportion of conspiracy and questionable sources to reliable sources being much higher. In comparison with the mainstream and health news, the right-leaning community is more influenced by the anti-vaccine and far-right communities, which is also reflected in the disparate vaccination rates in left and right U.S. states. The misinformation communities are also more vocal, either in vaccine or other discussions, relative to remaining communities, besides other behavioral differences.

cs.SI↗
arXiv2022

When Physics Meets Machine Learning: A Survey of Physics-Informed Machine Learning

Physics-informed machine learning (PIML), referring to the combination of prior knowledge of physics, which is the high level abstraction of natural phenomenons and human behaviours in the long history, with data-driven machine learning models, has emerged as an effective way to mitigate the shortage of training data, to increase models' generalizability and to ensure the physical plausibility of results. In this paper, we survey an abundant number of recent works in PIML and summarize them from three aspects: (1) motivations of PIML, (2) physics knowledge in PIML, (3) methods of physics knowledge integration in PIML. We also discuss current challenges and corresponding research opportunities in PIML.

cs.LG↗
arXiv2022

Construction of Large-Scale Misinformation Labeled Datasets from Social Media Discourse using Label Refinement

Malicious accounts spreading misinformation has led to widespread false and misleading narratives in recent times, especially during the COVID-19 pandemic, and social media platforms struggle to eliminate these contents rapidly. This is because adapting to new domains requires human intensive fact-checking that is slow and difficult to scale. To address this challenge, we propose to leverage news-source credibility labels as weak labels for social media posts and propose model-guided refinement of labels to construct large-scale, diverse misinformation labeled datasets in new domains. The weak labels can be inaccurate at the article or social media post level where the stance of the user does not align with the news source or article credibility. We propose a framework to use a detection model self-trained on the initial weak labels with uncertainty sampling based on entropy in predictions of the model to identify potentially inaccurate labels and correct for them using self-supervision or relabeling. The framework will incorporate social context of the post in terms of the community of its associated user for surfacing inaccurate labels towards building a large-scale dataset with minimum human effort. To provide labeled datasets with distinction of misleading narratives where information might be missing significant context or has inaccurate ancillary details, the proposed framework will use the few labeled samples as class prototypes to separate high confidence samples into false, unproven, mixture, mostly false, mostly true, true, and debunk information. The approach is demonstrated for providing a large-scale misinformation dataset on COVID-19 vaccines.

cs.SI↗
arXiv2022

On the Importance of Building High-quality Training Datasets for Neural Code Search

The performance of neural code search is significantly influenced by the quality of the training data from which the neural models are derived. A large corpus of high-quality query and code pairs is demanded to establish a precise mapping from the natural language to the programming language. Due to the limited availability, most widely-used code search datasets are established with compromise, such as using code comments as a replacement of queries. Our empirical study on a famous code search dataset reveals that over one-third of its queries contain noises that make them deviate from natural user queries. Models trained through noisy data are faced with severe performance degradation when applied in real-world scenarios. To improve the dataset quality and make the queries of its samples semantically identical to real user queries is critical for the practical usability of neural code search. In this paper, we propose a data cleaning framework consisting of two subsequent filters: a rule-based syntactic filter and a model-based semantic filter. This is the first framework that applies semantic query cleaning to code search datasets. Experimentally, we evaluated the effectiveness of our framework on two widely-used code search models and three manually-annotated code retrieval benchmarks. Training the popular DeepCS model with the filtered dataset from our framework improves its performance by 19.2% MRR and 21.3% Answer@1, on average with the three validation benchmarks.

cs.SE↗
arXiv2022

Classification of radial Kerr geodesic motion

We classify radial timelike geodesic motion of the exterior non-extremal Kerr spacetime by performing a taxonomy of inequivalent root structures of the first order radial geodesic equation using a novel compact notation and by implementing the constraints from polar, time and azimuthal motion. Four generic root structures with only simple roots give rise to eight non-generic root structures when either one root becomes coincident with the horizon, one root vanishes or two roots becomes coincident. We derive the explicit phase space of all such root systems in the basis of energy, angular momentum and Carter's constant and classify whether each corresponding radial geodesic motion is allowed or disallowed from existence of polar, time and azimuthal motion. The classification of radial motion within the ergoregion for both positive and negative energies leads to 6 distinguished values of the Kerr angular momentum. The classification of null radial motion and near-horizon extremal Kerr radial motion are obtained as limiting cases and compared with the literature. We explicitly parametrize the separatrix describing root systems with double roots as the union of the following three regions that are described by the same quartic respectively obtained when (1) the pericenter of bound motion becomes a double root; (2) the eccentricity of bound motion becomes zero; (3) the turning point of unbound motion becomes a double root.

gr-qc↗
arXiv2022

Breakdown of hydrodynamics from holographic pole collision

We study the breakdown of diffusive hydrodynamics in holographic systems dual to neutral dilatonic black holes with extremal near horizon geometries conformal to AdS$_2\times\,$R$^2$. We find that at low temperatures by tuning the effective gauge coupling constant in the infra-red, the lowest non-hydrodynamic mode, which collides with the charge diffusive mode and sets the scales at which diffusive hydrodynamics break down, could be either an infra-red mode or a slow mode, resulting in different scaling behaviors of the local equilibrium scales. We confirm that the upper bound for the charge diffusion constant is always satisfied using the velocity and timescale of local equilibration from the pole collision. We also examine the breakdown of hydrodynamics at general temperature and find that the convergence radius has nontrivial dependence on temperature, in addition to the effective gauge coupling constant.

hep-th↗
arXiv2022

Alternating current conductivity and superconducting properties of the holographic effective theory

We construct a holographic effective superconducting theory by considering a special gauge-axion higher derivative term. The gauge-axion coupling results in the transport behavior similar to the vortex response in the dual boundary field theory leading to non-Drude behavior of alternating current (AC) conductivity at the weak momentum dissipation. With the momentum dissipation increasing, a dip exhibits in the AC conductivity at low frequency. It is thought to be the result of a combination of the strong momentum dissipation and the gauge-axion coupling. In the superconducting phase, this gauge-axion coupling also plays a key role leading to a more evident gap at the low frequency conductivity. In addition, we also study the combined effects of the strength of momentum dissipation and various couplings among the gauge field, axion fields and the complex scalar field.

hep-th↗
arXiv2022

Box2Seg: Learning Semantics of 3D Point Clouds with Box-Level Supervision

Learning dense point-wise semantics from unstructured 3D point clouds with fewer labels, although a realistic problem, has been under-explored in literature. While existing weakly supervised methods can effectively learn semantics with only a small fraction of point-level annotations, we find that the vanilla bounding box-level annotation is also informative for semantic segmentation of large-scale 3D point clouds. In this paper, we introduce a neural architecture, termed Box2Seg, to learn point-level semantics of 3D point clouds with bounding box-level supervision. The key to our approach is to generate accurate pseudo labels by exploring the geometric and topological structure inside and outside each bounding box. Specifically, an attention-based self-training (AST) technique and Point Class Activation Mapping (PCAM) are utilized to estimate pseudo-labels. The network is further trained and refined with pseudo labels. Experiments on two large-scale benchmarks including S3DIS and ScanNet demonstrate the competitive performance of the proposed method. In particular, the proposed network can be trained with cheap, or even off-the-shelf bounding box-level annotations and subcloud-level tags.

cs.CV↗