arXiv subjects

Tao Xiang

Publications and source records attributed to Tao Xiang.

At least 325 records · Page 18Linked to original sources

arXiv2017

Recent Advances in Zero-shot Recognition

With the recent renaissance of deep convolution neural networks, encouraging breakthroughs have been achieved on the supervised recognition tasks, where each class has sufficient training data and fully annotated training data. However, to scale the recognition to a large number of classes with few or now training samples for each class remains an unsolved problem. One approach to scaling up the recognition is to develop models capable of recognizing unseen categories without any training instances, or zero-shot recognition/ learning. This article provides a comprehensive review of existing zero-shot recognition techniques covering various aspects ranging from representations of models, and from datasets and evaluation settings. We also overview related recognition tasks including one-shot and open set recognition which can be used as natural extensions of zero-shot recognition when limited number of class samples become available or when zero-shot recognition is implemented in a real-world setting. Importantly, we highlight the limitations of existing approaches and point out future research directions in this existing new research area.

cs.CV↗

arXiv2017

Majorana surface modes of nodal topological pairings in spin-$\frac{3}{2}$ semi-metals

Multi-component electronic systems can appear in solid state systems with active orbital band structures. They exhibit richer structures of topological superconductivity beyond the conventional scenarios of spin singlet and triplet pairings in spin-$\frac{1}{2}$ systems. Examples include the half-Heusler compounds RPtBi series (R for a rare earth element), whose electronic structures are described by the effective Luttinger-Kohn model with spin-$\frac{3}{2}$ fermions exhibiting strong spin-orbit coupling and band conversion. Recent experiments provide evidence to unconventional superconductivity in the YPtBi material with nodal spin-septet pairing. We systematically study topological pairing structures in spin-$\frac{3}{2}$ systems with cubic group symmetries and calculate surface Majorana spectra, which exhibit both the zero energy flat band and the cubic dispersion. The signatures of these surface states in the quasi-particle interference patterns are studied, which can be tested in future tunneling experiments.

cond-mat.supr-con↗

arXiv2017

Multi-scale Deep Learning Architectures for Person Re-identification

Person Re-identification (re-id) aims to match people across non-overlapping camera views in a public space. It is a challenging problem because many people captured in surveillance videos wear similar clothes. Consequently, the differences in their appearance are often subtle and only detectable at the right location and scales. Existing re-id models, particularly the recently proposed deep learning based ones match people at a single scale. In contrast, in this paper, a novel multi-scale deep learning model is proposed. Our model is able to learn deep discriminative feature representations at different scales and automatically determine the most suitable scales for matching. The importance of different spatial locations for extracting discriminative features is also learned explicitly. Experiments are carried out to demonstrate that the proposed model outperforms the state-of-the art on a number of benchmarks

cs.CV↗

arXiv2017

Superconductivity in pressurized CeRhGe3 and related non-centrosymmetric compounds

We report the discovery of superconductivity in pressurized CeRhGe3, until now the only remaining non-superconducting member of the isostructural family of non-centrosymmetric heavy-fermion compounds CeTX3 (T = Co, Rh, Ir and X = Si, Ge). Superconductivity appears in CeRhGe3 at a pressure of 19.6 GPa and the transition temperature Tc reaches a maximum value of 1.3 K at 21.5 GPa. This finding provides an opportunity to establish systematic correlations between superconductivity and materials properties within this family. Though ambient-pressure unit-cell volumes and critical pressures for superconductivity vary substantially across the series, all family members reach a maximum Tcmax at a common critical cell volume Vcrit, and Tcmax at Vcrit increases with increasing spin-orbit coupling strength of the d-electrons. These correlations show that substantial Kondo hybridization and spin-orbit coupling favor superconductivity in this family, the latter reflecting the role of broken centro-symmetry.

cond-mat.supr-con↗

arXiv2017

Weakly Supervised Image Annotation and Segmentation with Objects and Attributes

We propose to model complex visual scenes using a non-parametric Bayesian model learned from weakly labelled images abundant on media sharing sites such as Flickr. Given weak image-level annotations of objects and attributes without locations or associations between them, our model aims to learn the appearance of object and attribute classes as well as their association on each object instance. Once learned, given an image, our model can be deployed to tackle a number of vision problems in a joint and coherent manner, including recognising objects in the scene (automatic object annotation), describing objects using their attributes (attribute prediction and association), and localising and delineating the objects (object detection and semantic segmentation). This is achieved by developing a novel Weakly Supervised Markov Random Field Stacked Indian Buffet Process (WS-MRF-SIBP) that models objects and attributes as latent factors and explicitly captures their correlations within and across superpixels. Extensive experiments on benchmark datasets demonstrate that our weakly supervised model significantly outperforms weakly supervised alternatives and is often comparable with existing strongly supervised models on a variety of tasks including semantic segmentation, automatic image annotation and retrieval based on object-attribute associations.

cs.CV↗

arXiv2017

Zero-Shot Fine-Grained Classification by Deep Feature Learning with Semantics

Fine-grained image classification, which aims to distinguish images with subtle distinctions, is a challenging task due to two main issues: lack of sufficient training data for every class and difficulty in learning discriminative features for representation. In this paper, to address the two issues, we propose a two-phase framework for recognizing images from unseen fine-grained classes, i.e. zero-shot fine-grained classification. In the first feature learning phase, we finetune deep convolutional neural networks using hierarchical semantic structure among fine-grained classes to extract discriminative deep visual features. Meanwhile, a domain adaptation structure is induced into deep convolutional neural networks to avoid domain shift from training data to test data. In the second label inference phase, a semantic directed graph is constructed over attributes of fine-grained classes. Based on this graph, we develop a label propagation algorithm to infer the labels of images in the unseen classes. Experimental results on two benchmark datasets demonstrate that our model outperforms the state-of-the-art zero-shot learning models. In addition, the features obtained by our feature learning model also yield significant gains when they are used by other zero-shot learning models, which shows the flexility of our model in zero-shot fine-grained classification.

cs.CV↗

arXiv2017

Learning to Learn: Meta-Critic Networks for Sample Efficient Learning

We propose a novel and flexible approach to meta-learning for learning-to-learn from only a few examples. Our framework is motivated by actor-critic reinforcement learning, but can be applied to both reinforcement and supervised learning. The key idea is to learn a meta-critic: an action-value function neural network that learns to criticise any actor trying to solve any specified task. For supervised learning, this corresponds to the novel idea of a trainable task-parametrised loss generator. This meta-critic approach provides a route to knowledge transfer that can flexibly deal with few-shot and semi-supervised conditions for both reinforcement and supervised learning. Promising results are shown on both reinforcement and supervised learning problems.

cs.LG↗

arXiv2017

Analytic continuation with Padé decomposition

The ill-posed analytic continuation problem for Green's functions or self-energies can be done using the Padé rational polynomial approximation. However, to extract accurate results from this approximation, high precision input data of the Matsubara Green's function are needed. The calculation of the Matsubara Green's function generally involves a Matsubara frequency summation which cannot be evaluated analytically. Numerical summation is requisite but it converges slowly with the increase of the Matsubara frequency. Here we show that this slow convergence problem can be significantly improved by utilizing the Padé decomposition approach to replace the Matsubara frequency summation by a Padé frequency summation, and high precision input data can be obtained to successfully perform the Padé analytic continuation.

cond-mat.str-el↗

arXiv2017

Bayesian Joint Modelling for Object Localisation in Weakly Labelled Images

We address the problem of localisation of objects as bounding boxes in images and videos with weak labels. This weakly supervised object localisation problem has been tackled in the past using discriminative models where each object class is localised independently from other classes. In this paper, a novel framework based on Bayesian joint topic modelling is proposed, which differs significantly from the existing ones in that: (1) All foreground object classes are modelled jointly in a single generative model that encodes multiple object co-existence so that "explaining away" inference can resolve ambiguity and lead to better learning and localisation. (2) Image backgrounds are shared across classes to better learn varying surroundings and "push out" objects of interest. (3) Our model can be learned with a mixture of weakly labelled and unlabelled data, allowing the large volume of unlabelled images on the Internet to be exploited for learning. Moreover, the Bayesian formulation enables the exploitation of various types of prior knowledge to compensate for the limited supervision offered by weakly labelled data, as well as Bayesian domain adaptation for transfer learning. Extensive experiments on the PASCAL VOC, ImageNet and YouTube-Object videos datasets demonstrate the effectiveness of our Bayesian joint model for weakly supervised object localisation.

cs.CV↗

arXiv2017

The Electron Pairing of K$_x$Fe$_{2-y}$Se$_2$

We studied the pairing instabilities in K$_x$Fe$_{2-y}$Se$_2$ using a two stage functional renormalization group (FRG) method. Our results suggest the leading and subleading pairing symmetries are nodeless $d_{x^2-y^2}$ and nodal extended $s$ respectively. In addition, despite having no Fermi surfaces we find the buried hole bands make important contributions to the final effective interaction. From the bandstructure, spin susceptibility and the FRG results we conclude that the low energy effective interaction in K$_x$Fe$_{2-y}$Se$_2$ is well described by a $J_1-J_2$ model with dominant nearest-neighbor antiferromagnetic interaction $J_1$ (at least as far as the superconducting pairing is concerned). In the end we briefly mention several obvious experiments to test whether the pairing symmetry is indeed $d_{x^2-y^2}$.

cond-mat.supr-con↗

arXiv2017

Transferring a Semantic Representation for Person Re-Identification and Search

Learning semantic attributes for person re-identification and description-based person search has gained increasing interest due to attributes' great potential as a pose and view-invariant representation. However, existing attribute-centric approaches have thus far underperformed state-of-the-art conventional approaches. This is due to their non-scalable need for extensive domain (camera) specific annotation. In this paper we present a new semantic attribute learning approach for person re-identification and search. Our model is trained on existing fashion photography datasets -- either weakly or strongly labelled. It can then be transferred and adapted to provide a powerful semantic description of surveillance person detections, without requiring any surveillance domain supervision. The resulting representation is useful for both unsupervised and supervised person re-identification, achieving state-of-the-art and near state-of-the-art performance respectively. Furthermore, as a semantic representation it allows description-based person search to be integrated within the same framework.

cs.CV↗

arXiv2017

Phase transition of the q-state clock model: duality and tensor renormalization

We investigate the critical behavior and the duality property of the ferromagnetic $q$-state clock model on the square lattice based on the tensor-network formalism. From the entanglement spectra of local tensors defined in the original and dual lattices, we obtain the exact self-dual points for the model with $q \leq 5 $ and approximate self-dual points for $q \geq 6$. We calculate accurately the lower and upper critical temperatures for the six-state clock model from the fixed-point tensors determined using the higher-order tensor renormalization group method and compare with other numerical results.

cond-mat.stat-mech↗

arXiv2017

Deep Mutual Learning

Model distillation is an effective and widely used technique to transfer knowledge from a teacher to a student network. The typical application is to transfer from a powerful large network or ensemble to a small network, that is better suited to low-memory or fast execution requirements. In this paper, we present a deep mutual learning (DML) strategy where, rather than one way transfer between a static pre-defined teacher and a student, an ensemble of students learn collaboratively and teach each other throughout the training process. Our experiments show that a variety of network architectures benefit from mutual learning and achieve compelling results on CIFAR-100 recognition and Market-1501 person re-identification benchmarks. Surprisingly, it is revealed that no prior powerful teacher network is necessary -- mutual learning of a collection of simple student networks works, and moreover outperforms distillation from a more powerful yet static teacher.

cs.CV↗

arXiv2017

Cross-modal Subspace Learning for Fine-grained Sketch-based Image Retrieval

Sketch-based image retrieval (SBIR) is challenging due to the inherent domain-gap between sketch and photo. Compared with pixel-perfect depictions of photos, sketches are iconic renderings of the real world with highly abstract. Therefore, matching sketch and photo directly using low-level visual clues are unsufficient, since a common low-level subspace that traverses semantically across the two modalities is non-trivial to establish. Most existing SBIR studies do not directly tackle this cross-modal problem. This naturally motivates us to explore the effectiveness of cross-modal retrieval methods in SBIR, which have been applied in the image-text matching successfully. In this paper, we introduce and compare a series of state-of-the-art cross-modal subspace learning methods and benchmark them on two recently released fine-grained SBIR datasets. Through thorough examination of the experimental results, we have demonstrated that the subspace learning can effectively model the sketch-photo domain-gap. In addition we draw a few key insights to drive future research.

cs.CV↗

arXiv2017

Bayesian Joint Topic Modelling for Weakly Supervised Object Localisation

We address the problem of localisation of objects as bounding boxes in images with weak labels. This weakly supervised object localisation problem has been tackled in the past using discriminative models where each object class is localised independently from other classes. We propose a novel framework based on Bayesian joint topic modelling. Our framework has three distinctive advantages over previous works: (1) All object classes and image backgrounds are modelled jointly together in a single generative model so that "explaining away" inference can resolve ambiguity and lead to better learning and localisation. (2) The Bayesian formulation of the model enables easy integration of prior knowledge about object appearance to compensate for limited supervision. (3) Our model can be learned with a mixture of weakly labelled and unlabelled data, allowing the large volume of unlabelled images on the Internet to be exploited for learning. Extensive experiments on the challenging VOC dataset demonstrate that our approach outperforms the state-of-the-art competitors.

cs.CV↗

arXiv2017

Transfer Learning by Ranking for Weakly Supervised Object Annotation

Most existing approaches to training object detectors rely on fully supervised learning, which requires the tedious manual annotation of object location in a training set. Recently there has been an increasing interest in developing weakly supervised approach to detector training where the object location is not manually annotated but automatically determined based on binary (weak) labels indicating if a training image contains the object. This is a challenging problem because each image can contain many candidate object locations which partially overlaps the object of interest. Existing approaches focus on how to best utilise the binary labels for object location annotation. In this paper we propose to solve this problem from a very different perspective by casting it as a transfer learning problem. Specifically, we formulate a novel transfer learning based on learning to rank, which effectively transfers a model for automatic annotation of object location from an auxiliary dataset to a target dataset with completely unrelated object categories. We show that our approach outperforms existing state-of-the-art weakly supervised approach to annotating objects in the challenging VOC dataset.

cs.CV↗

arXiv2017

Semantic Autoencoder for Zero-Shot Learning

Existing zero-shot learning (ZSL) models typically learn a projection function from a feature space to a semantic embedding space (e.g.~attribute space). However, such a projection function is only concerned with predicting the training seen class semantic representation (e.g.~attribute prediction) or classification. When applied to test data, which in the context of ZSL contains different (unseen) classes without training data, a ZSL model typically suffers from the project domain shift problem. In this work, we present a novel solution to ZSL based on learning a Semantic AutoEncoder (SAE). Taking the encoder-decoder paradigm, an encoder aims to project a visual feature vector into the semantic space as in the existing ZSL models. However, the decoder exerts an additional constraint, that is, the projection/code must be able to reconstruct the original visual feature. We show that with this additional reconstruction constraint, the learned projection function from the seen classes is able to generalise better to the new unseen classes. Importantly, the encoder and decoder are linear and symmetric which enable us to develop an extremely efficient learning algorithm. Extensive experiments on six benchmark datasets demonstrate that the proposed SAE outperforms significantly the existing ZSL models with the additional benefit of lower computational cost. Furthermore, when the SAE is applied to supervised clustering problem, it also beats the state-of-the-art.

cs.CV↗

arXiv2017

Pair-checkerboard antiferromagnetic order in $β$-Fe$_4$Se$_5$ with $\sqrt{5}\times\sqrt{5}$ ordered Fe vacancies

The electronic structure of recently discovered $β$-Fe$_4$Se$_5$ with $\sqrt{5} \times \sqrt{5}$ ordered Fe vacancies is calculated using first-principles density functional theory. We find that the ground state is an antiferromagnetic (AFM) insulator in agreement with the experimental observation. In K$_2$Fe$_4$Se$_5$, it is known that the ground state is $\sqrt{5} \times \sqrt{5}$-blocked-checkerboard AFM ordered. But for this material, we find that the ground state is $\sqrt{5} \times \sqrt{5}$-pair-checkerboard AFM ordered, in which the intrablock four Fe spins exhibit the collinear AFM order and the interblock spins on any two nearest-neighboring sites are antiparallel aligned. This state is about 130 meV/Fe lower in energy than the $\sqrt{5} \times \sqrt{5}$-blocked-checkerboard AFM one. Electron doping can lower the energy of the $\sqrt{5} \times \sqrt{5}$-blocked-checkerboard AFM state and introduce a transition between these two states, suggesting that there is strong AFM fluctuation in FeSe-based materials upon doping. This provides a unified picture to understand the AFM orders in $β$-Fe$_4$Se$_5$ and in alkali-metal intercalated FeSe materials.

cond-mat.supr-con↗