arXiv subjects

Tian Xia

Publications and source records attributed to Tian Xia.

At least 109 records · Page 6Linked to original sources

arXiv2021

Pseudo-healthy synthesis with pathology disentanglement and adversarial learning

Pseudo-healthy synthesis is the task of creating a subject-specific `healthy' image from a pathological one. Such images can be helpful in tasks such as anomaly detection and understanding changes induced by pathology and disease. In this paper, we present a model that is encouraged to disentangle the information of pathology from what seems to be healthy. We disentangle what appears to be healthy and where disease is as a segmentation map, which are then recombined by a network to reconstruct the input disease image. We train our models adversarially using either paired or unpaired settings, where we pair disease images and maps when available. We quantitatively and subjectively, with a human study, evaluate the quality of pseudo-healthy images using several criteria. We show in a series of experiments, performed on ISLES, BraTS and Cam-CAN datasets, that our method is better than several baselines and methods from the literature. We also show that due to better training processes we could recover deformations, on surrounding tissue, caused by disease. Our implementation is publicly available at https://github.com/xiat0616/pseudo-healthy-synthesis. This paper has been accepted by Medical Image Analysis: https://doi.org/10.1016/j.media.2020.101719.

eess.IV↗

arXiv2020

Deep Multi-attribute Graph Representation Learning on Protein Structures

Graphs as a type of data structure have recently attracted significant attention. Representation learning of geometric graphs has achieved great success in many fields including molecular, social, and financial networks. It is natural to present proteins as graphs in which nodes represent the residues and edges represent the pairwise interactions between residues. However, 3D protein structures have rarely been studied as graphs directly. The challenges include: 1) Proteins are complex macromolecules composed of thousands of atoms making them much harder to model than micro-molecules. 2) Capturing the long-range pairwise relations for protein structure modeling remains under-explored. 3) Few studies have focused on learning the different attributes of proteins together. To address the above challenges, we propose a new graph neural network architecture to represent the proteins as 3D graphs and predict both distance geometric graph representation and dihedral geometric graph representation together. This gives a significant advantage because this network opens a new path from the sequence to structure. We conducted extensive experiments on four different datasets and demonstrated the effectiveness of the proposed method.

cs.LG↗

arXiv2020

AlignTTS: Efficient Feed-Forward Text-to-Speech System without Explicit Alignment

Targeting at both high efficiency and performance, we propose AlignTTS to predict the mel-spectrum in parallel. AlignTTS is based on a Feed-Forward Transformer which generates mel-spectrum from a sequence of characters, and the duration of each character is determined by a duration predictor.Instead of adopting the attention mechanism in Transformer TTS to align text to mel-spectrum, the alignment loss is presented to consider all possible alignments in training by use of dynamic programming. Experiments on the LJSpeech dataset show that our model achieves not only state-of-the-art performance which outperforms Transformer TTS by 0.03 in mean option score (MOS), but also a high efficiency which is more than 50 times faster than real-time.

eess.AS↗

arXiv2020

Learning Multi-Robot Decentralized Macro-Action-Based Policies via a Centralized Q-Net

In many real-world multi-robot tasks, high-quality solutions often require a team of robots to perform asynchronous actions under decentralized control. Decentralized multi-agent reinforcement learning methods have difficulty learning decentralized policies because of the environment appearing to be non-stationary due to other agents also learning at the same time. In this paper, we address this challenge by proposing a macro-action-based decentralized multi-agent double deep recurrent Q-net (MacDec-MADDRQN) which trains each decentralized Q-net using a centralized Q-net for action selection. A generalized version of MacDec-MADDRQN with two separate training environments, called Parallel-MacDec-MADDRQN, is also presented to leverage either centralized or decentralized exploration. The advantages and the practical nature of our methods are demonstrated by achieving near-centralized results in simulation and having real robots accomplish a warehouse tool delivery task in an efficient way.

cs.RO↗

arXiv2019

Plackett-Luce model for learning-to-rank task

List-wise based learning to rank methods are generally supposed to have better performance than point- and pair-wise based. However, in real-world applications, state-of-the-art systems are not from list-wise based camp. In this paper, we propose a new non-linear algorithm in the list-wise based framework called ListMLE, which uses the Plackett-Luce (PL) loss. Our experiments are conducted on the two largest publicly available real-world datasets, Yahoo challenge 2010 and Microsoft 30K. This is the first time in the single model level for a list-wise based system to match or overpass state-of-the-art systems in real-world datasets.

cs.IR↗

arXiv2019

A simple discriminative training method for machine translation with large-scale features

Margin infused relaxed algorithms (MIRAs) dominate model tuning in statistical machine translation in the case of large scale features, but also they are famous for the complexity in implementation. We introduce a new method, which regards an N-best list as a permutation and minimizes the Plackett-Luce loss of ground-truth permutations. Experiments with large-scale features demonstrate that, the new method is more robust than MERT; though it is only matchable with MIRAs, it has a comparatively advantage, easier to implement.

cs.CL↗

arXiv2019

Analysis of Regression Tree Fitting Algorithms in Learning to Rank

In learning to rank area, industry-level applications have been dominated by gradient boosting framework, which fits a tree using least square error principle. While in classification area, another tree fitting principle, weighted least square error, has been widely used, such as LogitBoost and its variants. However, there is a lack of analysis on the relationship between the two principles in the scenario of learning to rank. We propose a new principle named least objective loss based error that enables us to analyze the issue above as well as several important learning to rank models. We also implement two typical and strong systems and conduct our experiments in two real-world datasets. Experimental results show that our proposed method brings moderate improvements over least square error principle.

cs.IR↗

arXiv2019

RGB-T Image Saliency Detection via Collaborative Graph Learning

Image saliency detection is an active research topic in the community of computer vision and multimedia. Fusing complementary RGB and thermal infrared data has been proven to be effective for image saliency detection. In this paper, we propose an effective approach for RGB-T image saliency detection. Our approach relies on a novel collaborative graph learning algorithm. In particular, we take superpixels as graph nodes, and collaboratively use hierarchical deep features to jointly learn graph affinity and node saliency in a unified optimization framework. Moreover, we contribute a more challenging dataset for the purpose of RGB-T image saliency detection, which contains 1000 spatially aligned RGB-T image pairs and their ground truth annotations. Extensive experiments on the public dataset and the newly created dataset suggest that the proposed approach performs favorably against the state-of-the-art RGB-T saliency detection methods.

cs.CV↗

arXiv2019

Electrically conductive GNP/epoxy composites for out-of-autoclave thermoset curing through Joule heating

The development of scalable Out of Autoclave (OoA) in situ thermoset curing methods are required to overcome important drawbacks related to the autoclave based processing methods typically used in industry. The incorporation of graphene, an electrothermal carbon nanomaterial with the ability to transform electric energy into heat through Joule heating, emerges as a promising route to replace the conventional processing methods. In this work the electrical behaviour of both uncured and oven cured GNPs/epoxy composites with loadings of up to 10 wt% were evaluated and electrical percolation thresholds were established for both. Above the critical loading found for oven cured materials (8.5 wt%) the electrically conducting networks of GNPs formed in the matrix showed the ability to act as integrated nanoheaters when an electric current was passed through them, successfully curing the composites by Joule heating. Composites prepared by this OoA curing method (as an alternative to the traditional oven based one) at 10 wt% loading of GNPs were also prepared and compared to the oven cured ones. They showed more compact composite structures, with less microvoids and a preferred orientation of the GNPs in the matrix relative to the oven cured material at identical loading, as revealed by electron microscopy and polarized Raman spectroscopy, respectively. This microstructure and anisotropy induced by the electrically induced (i.e. OoA) cure led to GNPs/epoxy composites with superior electrical and mechanical properties (revealed by tensile testing). The well distributed GNP nanoparticles acting as nanoheaters integrated in a thermosetting matrix, in combination with excellent mechanical and electrical performances achieved for the overall graphene/epoxy composites and the simplicity associated to the method, should open the door to novel industrial applications.

physics.app-ph↗

arXiv2019

Adversarial Pseudo Healthy Synthesis Needs Pathology Factorization

Pseudo healthy synthesis, i.e. the creation of a subject-specific `healthy' image from a pathological one, could be helpful in tasks such as anomaly detection, understanding changes induced by pathology and disease or even as data augmentation. We treat this task as a factor decomposition problem: we aim to separate what appears to be healthy and where disease is (as a map). The two factors are then recombined (by a network) to reconstruct the input disease image. We train our models in an adversarial way using either paired or unpaired settings, where we pair disease images and maps (as segmentation masks) when available. We quantitatively evaluate the quality of pseudo healthy images. We show in a series of experiments, performed in ISLES and BraTS datasets, that our method is better than conditional GAN and CycleGAN, highlighting challenges in using adversarial methods in the image translation task of pseudo healthy image generation.

cs.CV↗

arXiv2018

Quantum states preparation of an atomic ensemble via cavity-assisted homodyne measurement

The quantum spin states of atomic ensemble are of special interesting for both fundamental studies and precision measurement applications. Here, we propose a scheme to prepare collective quantum states of an atomic ensemble placed in an optical cavity via homodyne measurement of probing light field. The effective interactions of atoms mediated by photons are enhanced by the optical cavity, and the output probe light could also be entangled with the collective spin states. By selectively measuring the quadrature of output light, we can prepare various quantum states, including superposition states of Dicke states and Dicke squeezed states. It is also demonstrated that the fidelity of prepared quantum state can be enhanced by repetitive homodyne detection and using longer probe laser pulses. Our scheme is feasible for experimental realization with current technologies, which may be used in future study of quantum mechanics and quantum metrology.

quant-ph↗

arXiv2017

Multi-View 3D Object Detection Network for Autonomous Driving

This paper aims at high-accuracy 3D object detection in autonomous driving scenario. We propose Multi-View 3D networks (MV3D), a sensory-fusion framework that takes both LIDAR point cloud and RGB images as input and predicts oriented 3D bounding boxes. We encode the sparse 3D point cloud with a compact multi-view representation. The network is composed of two subnetworks: one for 3D object proposal generation and another for multi-view feature fusion. The proposal network generates 3D candidate boxes efficiently from the bird's eye view representation of 3D point cloud. We design a deep fusion scheme to combine region-wise features from multiple views and enable interactions between intermediate layers of different paths. Experiments on the challenging KITTI benchmark show that our approach outperforms the state-of-the-art by around 25% and 30% AP on the tasks of 3D localization and 3D detection. In addition, for 2D detection, our approach obtains 10.3% higher AP than the state-of-the-art on the hard data among the LIDAR-based methods.

cs.CV↗

arXiv2017

Fully Convolutional Attention Networks for Fine-Grained Recognition

Fine-grained recognition is challenging due to its subtle local inter-class differences versus large intra-class variations such as poses. A key to address this problem is to localize discriminative parts to extract pose-invariant features. However, ground-truth part annotations can be expensive to acquire. Moreover, it is hard to define parts for many fine-grained classes. This work introduces Fully Convolutional Attention Networks (FCANs), a reinforcement learning framework to optimally glimpse local discriminative regions adaptive to different fine-grained domains. Compared to previous methods, our approach enjoys three advantages: 1) the weakly-supervised reinforcement learning procedure requires no expensive part annotations; 2) the fully-convolutional architecture speeds up both training and testing; 3) the greedy reward strategy accelerates the convergence of the learning. We demonstrate the effectiveness of our method with extensive experiments on four challenging fine-grained benchmark datasets, including CUB-200-2011, Stanford Dogs, Stanford Cars and Food-101.

cs.CV↗

arXiv2016

Vehicle Detection from 3D Lidar Using Fully Convolutional Network

Convolutional network techniques have recently achieved great success in vision based detection tasks. This paper introduces the recent development of our research on transplanting the fully convolutional network technique to the detection tasks on 3D range scan data. Specifically, the scenario is set as the vehicle detection task from the range data of Velodyne 64E lidar. We proposes to present the data in a 2D point map and use a single 2D end-to-end fully convolutional network to predict the objectness confidence and the bounding boxes simultaneously. By carefully design the bounding box encoding, it is able to predict full 3D bounding boxes even using a 2D convolutional network. Experiments on the KITTI dataset shows the state-of-the-art performance of the proposed method.

cs.CV↗

arXiv2016

Output-Feedback Synchronizability of Linear Time-Invariant Systems

The paper studies the output-feedback synchronization problem for a network of identical, linear time-invariant systems. A criterion to test network synchronization is derived and the class of output-feedback synchronizable systems is introduced and characterized by sufficient and necessary conditions. In particular it is observed that output-feedback stabilizability is sufficient but not necessary for output-feedback synchronizability. In the special case of single-input single-output systems, conditions are derived in the frequency domain. The theory is illustrated with several examples.

math.OC↗

arXiv2016

A Necessary and Sufficient Condition for Having Independent TE and TM Modes in an Anisotropic Waveguide

Whether there exist independent transverse electric (TE) and transverse magnetic (TM) modes in a metallic waveguide filled with an anisotropic medium is a fundamental question in electromagnetics waveguide theory, but so far no definitive answers have been published. This paper establishes a necessary and sufficient condition for having independent TE and TM modes in a waveguide filled with a homogeneous lossless anisotropic medium based on both waveguide theory in electromagnetics and basic knowledge in mathematics. Moreover, for the independent TE modes, we prove the propagation constants obtained from both the longitudinal scalar magnetic field stimulation and the transverse vector electric field stimulation are the same; for the independent TM modes, the propagation constants obtained from both the longitudinal scalar electric field stimulation and the transverse vector magnetic field stimulation are the same. This necessary and sufficient condition is a new theoretical result in electromagnetic waveguide theory, and is expected to be valuable for the design of waveguides filled with anisotropic media.

physics.class-ph↗

arXiv2016

DeepPicker: a Deep Learning Approach for Fully Automated Particle Picking in Cryo-EM

Particle picking is a time-consuming step in single-particle analysis and often requires significant interventions from users, which has become a bottleneck for future automated electron cryo-microscopy (cryo-EM). Here we report a deep learning framework, called DeepPicker, to address this problem and fill the current gaps toward a fully automated cryo-EM pipeline. DeepPicker employs a novel cross-molecule training strategy to capture common features of particles from previously-analyzed micrographs, and thus does not require any human intervention during particle picking. Tests on the recently-published cryo-EM data of three complexes have demonstrated that our deep learning based scheme can successfully accomplish the human-level particle picking process and identify a sufficient number of particles that are comparable to those manually by human experts. These results indicate that DeepPicker can provide a practically useful tool to significantly reduce the time and manual effort spent in single-particle analysis and thus greatly facilitate high-resolution cryo-EM structure determination.

q-bio.QM↗

arXiv2015

A Bayesian regression tree approach to identify the effect of nanoparticles' properties on toxicity profiles

We introduce a Bayesian multiple regression tree model to characterize relationships between physico-chemical properties of nanoparticles and their in-vitro toxicity over multiple doses and times of exposure. Unlike conventional models that rely on data summaries, our model solves the low sample size issue and avoids arbitrary loss of information by combining all measurements from a general exposure experiment across doses, times of exposure, and replicates. The proposed technique integrates Bayesian trees for modeling threshold effects and interactions, and penalized B-splines for dose- and time-response surface smoothing. The resulting posterior distribution is sampled by Markov Chain Monte Carlo. This method allows for inference on a number of quantities of potential interest to substantive nanotoxicology, such as the importance of physico-chemical properties and their marginal effect on toxicity. We illustrate the application of our method to the analysis of a library of 24 nano metal oxides.

stat.AP↗