Zhen Zhang, Qiang Li
Jul 10, 2020·q-bio.PE·PDF Despite its historical and biological stability, the sex ratio at birth (SRB) has risen in parts of the world in the last several decades. The resultant demographic consequences, mostly on sex imbalance, are well documented, typically including "missing girls/women" and "marriage squeeze." However, the SRB-induced impact on demographic dynamics, particularly its underlying mechanism, has not been explored in depth. We aim to investigate the impact of the SRB rise on the size, structure, and growth of a population, particularly emphasizing on population aging. We provide a simple framework, derived from classical stable population models, to analyze how the SRB rise can reduce the population size and make the population old. We demonstrate that the cohorts born with a higher SRB are smaller in size than those with a lower SRB. As the affected cohorts are born into the population, their smaller size will reduce the total population size, thereby lifting the fraction of old people that were born with the original SRB and have the same size as before. The resultant population aging speed increases as the cohorts with the new SRB take an increasing share of the population. This study adds that, in addition to fertility and mortality, the SRB can be a driving factor of population dynamics, especially when it moves far above normal biological levels.
Zhen Zhang, Jing-Yang You, Bo Gu, Gang Su
Two-dimensional (2D) Janus semiconductors with mirror asymmetry can introduce novel properties, such as large spin-orbit coupling (SOC) and normal piezoelectric polarization, which have attracted a great interest for their potential applications. Inspired by the recently fabricated 2D ferromagnetic (FM) semiconductor CrI3, a stable 2D (in x-y plane) antiferromagnetic (AFM) Janus semiconductor Fe2Cl3I3 with normal sublattice magnetization (m//z) is obtained by density functional theory calculations. By applying a tensile strain, the four magnetic states sequentially occur: AFM with m//z of sublattice, AFM with m//xy of sublattice, FM with m//xy, and FM with m//z. Such novel magnetic phase diagram driven by strain can be well understood by the spin-spin interactions including the third nearest-neighbor hoppings with the single-ion anisotropy, in which the SOC of I atoms is found to play an essential role. In addition, the electric polarization of Fe2Cl3I3 preserves with strain due to the broken inversion symmetry. Our results predict the rare Janus material Fe2Cl3I3 as an example of 2D semiconductors with both spin and charge polarizations, and reveal the highly sensitive strain-controlled magnetic states and magnetization direction, which highlights the 2D magnetic Janus semiconductor as a new platform to design spintronic materials.
Pengzhan Jin, Zhen Zhang, Aiqing Zhu, Yifa Tang, George Em Karniadakis
We propose new symplectic networks (SympNets) for identifying Hamiltonian systems from data based on a composition of linear, activation and gradient modules. In particular, we define two classes of SympNets: the LA-SympNets composed of linear and activation modules, and the G-SympNets composed of gradient modules. Correspondingly, we prove two new universal approximation theorems that demonstrate that SympNets can approximate arbitrary symplectic maps based on appropriate activation functions. We then perform several experiments including the pendulum, double pendulum and three-body problems to investigate the expressivity and the generalization ability of SympNets. The simulation results show that even very small size SympNets can generalize well, and are able to handle both separable and non-separable Hamiltonian systems with data points resulting from short or long time steps. In all the test cases, SympNets outperform the baseline models, and are much faster in training and prediction. We also develop an extended version of SympNets to learn the dynamics from irregularly sampled data. This extended version of SympNets can be thought of as a universal model representing the solution to an arbitrary Hamiltonian system.
Zhen Zhang, Dong-Bo Zhang, Tao Sun, Renata M. Wentzcovitch
Knowledge of lattice anharmonicity is essential to elucidate distinctive thermal properties in crystalline solids. Yet, accurate \textit{ab initio} investigations of lattice anharmonicity encounter difficulties owing to the cumbersome computations. Here we introduce the phonon quasiparticle approach and review its application to various materials. This method efficiently and reliably addresses lattice anharmonicity by combining \textit{ab initio} molecular dynamics and lattice dynamics calculations. Thus, in principle, it accounts for full anharmonic effects and overcomes finite-size effects typical of \textit{ab initio} molecular dynamics. The validity and effectiveness of the current approach are demonstrated in the computation of thermodynamic and heat transport properties of weakly and strongly anharmonic systems.
Xiao-Yun Zhao, Shao-Lin Xiong, Xiang-Yang Wen, Xin-Qiao Li, Ce Cai, Shuo Xiao, Qi Luo, Wen-Xi Peng, Dong-Ya Guo, Zheng-Hua An, Ke Gong, Jin-Yuan Liao, Yan-Qiu Zhang, Yue Huang, Lu Li, Xing Wen, Fei Zhang, Jing Duan, Chen-Wei Wang, Dong-Li Shi, Peng Zhang, Qi-Bin Yi, Chao-Yang Li, Yan-Bing Xu, Xiao-Hua Liang, Ya-Qing Liu, Da-Li Zhang, Xi-Lei Sun, Fan Zhang, Gang Chen, Huan-Yu Wang, Sheng Yang, Xiao-Jing Liu, Min Gao, Mao-Shun Li, Jin-Zhou Wang, Xing Zhou, Yi Zhao, Wang-Chen Xue, Chao Zheng, Jia-Cong Liu, Xing-Bo Han, Jin-Ling Qi, Jia Huang, Ke-Ke Zhang, Can Chen, Xiong-Tao Yang, Dong-Jie Hou, Yu-Sa Wang, Rui Qiao, Xiang Ma, Xiao-Bo Li, Ping Wang, Xin-Ying Song, Li-Ming Song, Shi-Jie Zheng, Bing Li, Hong-Mei Zhang, Yue Zhu, Wei Chen, Jian-Jian He, Zhen Zhang, Jin Hou, Hong-Jun Wang, Yan-Chao Hao, Xiang-Yu Wang, Zong-Yuan Yang, Zhi-Long Wen, Zhi Chang, Yuan-Yuan Du, Rui Gao, Xiao-Fei Lan, Yan-Guo Li, Gang Li, Xu-Fang Li, Fang-Jun Lu, Hong Lu, Bin Meng, Feng Shi, Hui Wang, Hui-Zhen Wang, Yu-Peng Xu, Jia-Wei Yang, Xue-Juan Yang, Shuang-Nan Zhang, Chao-Yue Zhang, Cheng-Mo Zhang, Zhi-Cheng Tang, Cheng Cheng
Realtime trigger and localization of bursts are the key functions of GECAM, which is an all-sky gamma-ray monitor launched in Dec 10, 2020. We developed a multifunctional trigger and localization software operating on the CPU of the GECAM electronic box (EBOX). This onboard software has the following features: high trigger efficiency for real celestial bursts with a suppression of false triggers caused by charged particle bursts and background fluctuation, dedicated localization algorithm optimized for short and long bursts respetively, short time latency of the trigger information which is downlinked throught the BeiDou satellite navigation System (BDS). This paper presents the detailed design and deveopment of this trigger and localization software system of GECAM, including the main functions, general design, workflow and algorithms, as well as the verification and demonstration of this software, including the on-ground trigger tests with simulated gamma-ray bursts made by a dedicated X-ray tube and the in-flight performance to real gamma-ray bursts and magnetar bursts.
Zhen Zhang
The mysterious dark energy remains one of the greatest puzzles of modern science. Current detections for it are mostly indirect. The spacetime effects of dark energy can be locally described by the SdSw metric. Understanding these local effects exactly is an essential step towards the direct probe of dark energy. From first principles, we prove that dark energy can exert a repulsive dark force on astrophysical scales, different from the Newtonian attraction of both visible and dark matter. One way of measuring local effects of dark energy is through the gravitational deflection of light. We geometrize the bending of light in any curved static spacetime. First of all, we define a generalized deflection angle, referred to as the Gaussian deflection angle, in a mathematically strict and conceptually clean way. Basing on the Gauss-Bonnet theorem, we then prove that the Gaussian deflection angle is equivalent to the surface integral of the Gaussian curvature over a chosen lensing patch. As an application of the geometrization, we study the problem of whether dark energy affects the bending of light and provide a strict solution to this problem in the SdSw spacetime. According to this solution, we propose a method to overcome the difficulty of measuring local dark energy effects. Exactly speaking, we find that the lensing effect of dark energy can be enhanced by 14 orders of magnitude when properly choosing the lensing patch in certain cases. It means that we can probe the existence and nature of dark energy directly in our Solar System. This points to an exciting direction to help unraveling the great mystery of dark energy.
Wei Jiang, Zhen Zhang, Zeyu Zhou
By introducing height dependency in the surface energy density, we propose a novel regularized variational model to simulate wetting/dewetting problems. The regularized model leads to the appearance of a precursor layer which covers the bare substrate, with the precursor height depending on the regularization parameter $\varepsilon$. The new model enjoys lots of advantages in analysis and imulations. With the help of the precursor layer, the regularized model is naturally extended to a larger domain than that of the classical sharp-interface model, and thus can be solved in a fixed domain. There is no need to explicitly track the contact line motion, and difficulties arising from free boundary problems can be avoided. In addition, topological change events can be automatically captured. Under some mild and physically meaningful conditions, we show the positivity-preserving property of the minimizers of the new model. By using asymptotic analysis and $Γ$-convergence, we investigate the convergence relations between the new regularized model and the classical sharp-interface model. Finally, numerical results are provided to validate our theoretical analysis, as well as the accuracy and efficiency of the new regularized model.
Ce Cai, Wangchen Xue, Chengkui Li, Shaolin Xiong, Shuangnan Zhang, Lin Lin, Xiaobo Li, Mingyu Ge, Haisheng Zhao, Liming Song, Fangjun Lu, Shu Zhang, Yanqiu Zhang, Shuo Xiao, Youli Tuo, Qibin Yi, Zhiwei Guo, Shenglun Xie, Yi Zhao, Zhen Zhang, Qingxin Li, Jiacong Liu, Chao Zheng, Ping Wang
Mar 31, 2022·astro-ph.HE·PDF Magnetars are neutron stars with extreme magnetic field and sometimes manifest as soft gamma-ray repeaters (SGRs). SGR J1935+2154 is one of the most prolific bursters and the first confirmed source of fast radio burst (i.e. FRB 200428). Encouraged by the discovery of the first X-ray counterpart of FRB, Insight-Hard X-ray Modulation Telescope (Insight-HXMT) implemented a dedicated 33-day long ToO observation of SGR J1935+2154 since April 28, 2020. With the HE, ME, and LE telescopes, Insight-HXMT provides a thorough monitoring of burst activity evolution of SGR J1935+2154, in a very broad energy range (1-250 keV) with high temporal resolution and high sensitivity, resulting in a unique valuable data set for detailed studies of SGR J1935+2154. In this work, we conduct a comprehensive analysis of this observation including detailed burst search, identification and temporal analyses. After carefully removing false triggers, we find a total of 75 bursts from SGR J1935+2154, out of which 70 are single-pulsed. The maximum burst rate is about 56 bursts/day. Both the burst duration and the waiting time between two successive bursts follow log-normal distributions, consistent with previous studies. We also find that bursts with longer duration (some are multi-pulsed) tend to occur during the period with relatively high burst rate. There is no correlation between the waiting time and the fluence or duration of either the former or latter burst. It also seems that there is no correlation between burst duration and hardness ratio, in contrast to some previous reports. In addition, we do not find any X-ray burst associated with any reported radio bursts except for FRB 200428.
Zhen Zhang, Ignavier Ng, Dong Gong, Yuhang Liu, Ehsan M Abbasnejad, Mingming Gong, Kun Zhang, Javen Qinfeng Shi
Recovering underlying Directed Acyclic Graph (DAG) structures from observational data is highly challenging due to the combinatorial nature of the DAG-constrained optimization problem. Recently, DAG learning has been cast as a continuous optimization problem by characterizing the DAG constraint as a smooth equality one, generally based on polynomials over adjacency matrices. Existing methods place very small coefficients on high-order polynomial terms for stabilization, since they argue that large coefficients on the higher-order terms are harmful due to numeric exploding. On the contrary, we discover that large coefficients on higher-order terms are beneficial for DAG learning, when the spectral radiuses of the adjacency matrices are small, and that larger coefficients for higher-order terms can approximate the DAG constraints much better than the small counterparts. Based on this, we propose a novel DAG learning method with efficient truncated matrix power iteration to approximate geometric series based DAG constraints. Empirically, our DAG learning method outperforms the previous state-of-the-arts in various settings, often by a factor of $3$ or more in terms of structural Hamming distance.
Zhen Zhang, Luming Duan
Jun 30, 2014·quant-ph·PDF We introduce a new class of quantum many-particle entangled states, called the Dicke squeezed (or DS) states, which can be used to improve the precision in quantum metrology beyond the standard quantum limit. We show that the enhancement in measurement precision is characterized by a single experimentally detectable parameter, called the Dicke squeezing $ξ_{D}$, which also bounds the entanglement depth for this class of states. The measurement precision approaches the ultimate Heisenberg limit as $ξ_{D}$ attains the minimum in an ideal Dicke state. Compared with other entangled states, we show that the Dicke squeezed states are more robust to decoherence and give better measurement precision under typical experimental noise.
Yan-Lin Tang, Hua-Lei Yin, Si-Jing Chen, Yang Liu, Wei-Jun Zhang, Xiao Jiang, Lu Zhang, Jian Wang, Li-Xing You, Jian-Yu Guan, Dong-Xu Yang, Zhen Wang, Hao Liang, Zhen Zhang, Nan Zhou, Xiongfeng Ma, Teng-Yun Chen, Qiang Zhang, Jian-Wei Pan
Aug 11, 2014·quant-ph·PDF A main type of obstacles of practical applications of quantum key distribution (QKD) network is various attacks on detection. Measurement-device-independent QKD (MDIQKD) protocol is immune to all these attacks and thus a strong candidate for network security. Recently, several proof-of-principle demonstrations of MDIQKD have been performed. Although novel, those experiments are implemented in the laboratory with secure key rates less than 0.1 bps. Besides, they need manual calibration frequently to maintain the system performance. These aspects render these demonstrations far from practicability. Thus, justification is extremely crucial for practical deployment into the field environment. Here, by developing an automatic feedback MDIQKD system operated at a high clock rate, we perform a field test via deployed fiber network of 30 km total length, achieving a 16.9 bps secure key rate. The result lays the foundation for a global quantum network which can shield from all the detection-side attacks.
Zhen Zhang, Che Ming Ko
Within the framework of the relativistic Vlasov-Uehling-Uhlenbeck transport model based on the relativistic nonlinear NL$ρ$ interaction, we study pion in-medium effects on the $π^-/π^+$ ratio in Au+Au collisions at the energy of $E/A=400~\mathrm{MeV}$. These effects include the isospin-dependent pion s-wave and p-wave potentials, which are taken from calculations based on the chiral perturbation theory and the $Δ$-hole model, respectively. We find that the $π^-/π^+$ ratio in this collision is suppressed by the pion s-wave potential but enhanced by the p-wave potential, with a net effect of a significantly suppressed $π^-/π^+$ ratio. Including also the in-medium threshold effects on $Δ$ resonance production and decay and using a nuclear symmetry energy with a slope parameter $L=59~\mathrm{MeV}$ by reducing the coupling of isovector-vector $ρ$ meson to nucleon, our result is in good agreement with measured $π^-/π^+$ ratio from the FOPI Collaboration. We further investigate the pion in-medium effects on the ratio of charged pions as a function of their kinetic energies.
Zhen Zhang
Let $A$ be a finite dimensional algebra over an algebraically closed field. We present a relationship between simple-minded systems and coherent rings.
Chen-Wei Wang, Wen-Jun Tan, Shao-Lin Xiong, Shu-Xu Yi, Rahim Moradi, Bing Li, Zhen Zhang, Yu Wang, Yan-Zhi Meng, Jia-Cong Liu, Yue Wang, Sheng-Lun Xie, Wang-Chen Xue, Zheng-Hang Yu, Peng Zhang, Wen-Long Zhang, Yan-Qiu Zhang, Chao Zheng
Type I gamma-ray bursts (GRBs) are believed to originate from compact binary merger usually with duration less than 2 seconds for the main emission. However, recent observations of GRB 211211A and GRB 230307A indicate that some merger-origin GRBs could last much longer. Since they show strikingly similar properties (indicating a common mechanism) which are different from the classic "long"-short burst (e.g. GRB 060614), forming an interesting subclass of type I GRBs, we suggest to name them as type IL GRBs. By identifying the first peak of GRB 230307A as a quasi-thermal precursor, we find that the prompt emission of type IL GRB is composed of three episodes: (1) a precursor followed by a short quiescent (or weak emission) period, (2) a long-duration main emission, and (3) an extended emission. With this burst pattern, a good candidate, GRB 170228A, was found in the Fermi/GBM archive data, and subsequent temporal and spectral analyses indeed show that GRB 170228A falls in the same cluster with GRB 211211A and GRB 230307A in many diagnostic figures. Thus this burst pattern could be a good reference for rapidly identifying type IL GRB and conducting low-latency follow-up observation. We estimated the occurrence rate and discussed the physical origins and implications for the three emission episodes of type IL GRBs. Our analysis suggests the pre-merger precursor model, especially the super flare model, is more favored for type IL GRBs.
Zhen Zhang, Bingsheng He
Unsupervised graph domain adaptation (UGDA) focuses on transferring knowledge from labeled source graph to unlabeled target graph under domain discrepancies. Most existing UGDA methods are designed to adapt information from a single source domain, which cannot effectively exploit the complementary knowledge from multiple source domains. Furthermore, their assumptions that the labeled source graphs are accessible throughout the training procedure might not be practical due to privacy, regulation, and storage concerns. In this paper, we investigate multi-source-free unsupervised graph domain adaptation, i.e., adapting knowledge from multiple source domains to an unlabeled target domain without utilizing labeled source graphs but relying solely on source pre-trained models. Unlike previous multi-source domain adaptation approaches that aggregate predictions at model level, we introduce a novel model named GraphATA which conducts adaptation at node granularity. Specifically, we parameterize each node with its own graph convolutional matrix by automatically aggregating weight matrices from multiple source models according to its local context, thus realizing dynamic adaptation over graph structured data. We also demonstrate the capability of GraphATA to generalize to both model-centric and layer-centric methods. Comprehensive experiments on various public datasets show that our GraphATA can consistently surpass recent state-of-the-art baselines with different gains.
Zhuolin Li, Zhen Zhang, Witold Pedrycz
This paper introduces a novel incremental preference elicitation-based approach to learning potentially non-monotonic preferences in multi-criteria sorting (MCS) problems, enabling decision makers to progressively provide assignment example preference information. Specifically, we first construct a max-margin optimization-based model to model potentially non-monotonic preferences and inconsistent assignment example preference information in each iteration of the incremental preference elicitation process. Using the optimal objective function value of the max-margin optimization-based model, we devise information amount measurement methods and question selection strategies to pinpoint the most informative alternative in each iteration within the framework of uncertainty sampling in active learning. Once the termination criterion is satisfied, the sorting result for non-reference alternatives can be determined through the use of two optimization models, i.e., the max-margin optimization-based model and the complexity controlling optimization model. Subsequently, two incremental preference elicitation-based algorithms are developed to learn potentially non-monotonic preferences, considering different termination criteria. Ultimately, we apply the proposed approach to a credit rating problem to elucidate the detailed implementation steps, and perform computational experiments on both artificial and real-world data sets to compare the proposed question selection strategies with several benchmark strategies.
Zheng-Hua An, S. Antier, Xing-Zi Bi, Qing-Cui Bu, Ce Cai, Xue-Lei Cao, Anna-Elisa Camisasca, Zhi Chang, Gang Chen, Li Chen, Tian-Xiang Chen, Wen Chen, Yi-Bao Chen, Yong Chen, Yu-Peng Chen, Michael W. Coughlin, Wei-Wei Cui, Zi-Gao Dai, T. Hussenot-Desenonges, Yan-Qi Du, Yuan-Yuan Du, Yun-Fei Du, Cheng-Cheng Fan, Filippo Frontera, He Gao, Min Gao, Ming-Yu Ge, Ke Gong, Yu-Dong Gu, Ju Guan, Dong-Ya Guo, Zhi-Wei Guo, Cristiano Guidorzi, Da-Wei Han, Jian-Jian He, Jun-Wang He, Dong-Jie Hou, Yue Huang, Jia Huo, Zhen Ji, Shu-Mei Jia, Wei-Chun Jiang, David Alexander Kann, A. Klotz, Ling-Da Kong, Lin Lan, An Li, Bing Li, Chao-Yang Li, Cheng-Kui Li, Gang Li, Mao-Shun Li, Ti-Pei Li, Wei Li, Xiao-Bo Li, Xin-Qiao Li, Xu-Fang Li, Yan-Guo Li, Zheng-Wei Li, Jing Liang, Xiao-Hua Liang, Jin-Yuan Liao, Lin Lin, Cong-Zhan Liu, He-Xin Liu, Hong-Wei Liu, Jia-Cong Liu, Xiao-Jing Liu, Ya-Qing Liu, Yu-Rong Liu, Fang-Jun Lu, Hong Lu, Xue-Feng Lu, Qi Luo, Tao Luo, Bin-Yuan Ma, Fu-Li Ma, Rui-Can Ma, Xiang Ma, Romain Maccary, Ji-Rong Mao, Bin Meng, Jian-Yin Nie, Mauro Orlandini, Ge Ou, Jing-Qiang Peng, Wen-Xi Peng, Rui Qiao, Jin-Lu Qu, Xiao-Qin Ren, Jing-Yan Shi, Qi Shi, Li-Ming Song, Xin-Ying Song, Ju Su, Gong-Xing Sun, Liang Sun, Xi-Lei Sun, Wen-Jun Tan, Ying Tan, Lian Tao, You-Li Tuo, Damien Turpin, Jin-Zhou Wang, Chen Wang, Chen-Wei Wang, Hong-Jun Wang, Hui Wang, Jin Wang, Ling-Jun Wang, Peng-Ju Wang, Ping Wang, Wen-Shuai Wang, Xiang-Yu Wang, Xi-Lu Wang, Yu-Sa Wang, Yue Wang, Xiang-Yang Wen, Bo-Bing Wu, Bai-Yang Wu, Hong Wu, Sheng-Hui Xiao, Shuo Xiao, Yun-Xiang Xiao, Sheng-Lun Xie, Shao-Lin Xiong, Sen-Lin Xiong, Dong Xu, He Xu, Yan-Jun Xu, Yan-Bing Xu, Ying-Chen Xu, Yu-Peng Xu, Wang-Chen Xue, Sheng Yang, Yan-Ji Yang, Zi-Xu Yang, Wen-Tao Ye, Qi-Bin Yi, Shu-Xu Yi, Qian-Qing Yin, Yuan You, Yun-Wei Yu, Wei Yu, Wen-Hui Yu, Ming Zeng, Bing Zhang, Bin-Bin Zhang, Da-Li Zhang, Fan Zhang, Hong-Mei Zhang, Juan Zhang, Liang Zhang, Peng Zhang, Peng Zhang, Shu Zhang, Shuang-Nan Zhang, Wan-Chang Zhang, Xiao-Feng Zhang, Xiao-Lu Zhang, Yan-Qiu Zhang, Yan-Ting Zhang, Yi-Fei Zhang, Yuan-Hang Zhang, Zhen Zhang, Guo-Ying Zhao, Hai-Sheng Zhao, Hong-Yu Zhao, Qing-Xia Zhao, Shu-Jie Zhao, Xiao-Yun Zhao, Xiao-Fan Zhao, Yi Zhao, Chao Zheng, Shi-Jie Zheng, Deng-Ke Zhou, Xing Zhou, Xiao-Cheng Zhu
Robert Connelly, Bill Jackson, Shin-ichi Tanigawa, Zhen Zhang
Here we propose a class of frameworks in the plane, braced polygons, that may be globally rigid and are analogous to convex polyopes in 3 space that are rigid by Cauchy's rigidity Theorem in 1813.
Zhen Zhang, Xiangyu Chu, Yunxi Tang, Lulu Zhao, Jing Huang, Zhongliang Jiang, K. W. Samuel Au
Manipulating elasto-plastic objects remains a significant challenge due to severe self-occlusion, difficulties of representation, and complicated dynamics. This work proposes a novel framework for elasto-plastic object manipulation with a quasi-static assumption for motions, leveraging 3D occupancy to represent such objects, a learned dynamics model trained with 3D occupancy, and a learning-based predictive control algorithm to address these challenges effectively. We build a novel data collection platform to collect full spatial information and propose a pipeline for generating a 3D occupancy dataset. To infer the 3D occupancy during manipulation, an occupancy prediction network is trained with multiple RGB images supervised by the generated dataset. We design a deep neural network empowered by a 3D convolution neural network (CNN) and a graph neural network (GNN) to predict the complex deformation with the inferred 3D occupancy results. A learning-based predictive control algorithm is introduced to plan the robot actions, incorporating a novel shape-based action initialization module specifically designed to improve the planner efficiency. The proposed framework in this paper can successfully shape the elasto-plastic objects into a given goal shape and has been verified in various experiments both in simulation and the real world.
Yuhang Liu, Zhen Zhang, Dong Gong, Mingming Gong, Biwei Huang, Anton van den Hengel, Kun Zhang, Javen Qinfeng Shi
Multi-source domain adaptation (MSDA) addresses the challenge of learning a label prediction function for an unlabeled target domain by leveraging both the labeled data from multiple source domains and the unlabeled data from the target domain. Conventional MSDA approaches often rely on covariate shift or conditional shift paradigms, which assume a consistent label distribution across domains. However, this assumption proves limiting in practical scenarios where label distributions do vary across domains, diminishing its applicability in real-world settings. For example, animals from different regions exhibit diverse characteristics due to varying diets and genetics. Motivated by this, we propose a novel paradigm called latent covariate shift (LCS), which introduces significantly greater variability and adaptability across domains. Notably, it provides a theoretical assurance for recovering the latent cause of the label variable, which we refer to as the latent content variable. Within this new paradigm, we present an intricate causal generative model by introducing latent noises across domains, along with a latent content variable and a latent style variable to achieve more nuanced rendering of observational data. We demonstrate that the latent content variable can be identified up to block identifiability due to its versatile yet distinct causal structure. We anchor our theoretical insights into a novel MSDA method, which learns the label distribution conditioned on the identifiable latent content variable, thereby accommodating more substantial distribution shifts. The proposed approach showcases exceptional performance and efficacy on both simulated and real-world datasets.