Mutual Constraint Research Articles

Effect of global constraint on the mechanical behavior of gradient materials

The gradient structured (GS) samples studied here consists of gradient layer (s) with grain size gradient and a coarse-grained (CG) layer. Gradient materials have been found to have superior mechanical properties due to the hetero-deformation induced (HDI) strengthening and strain hardening. However, the effect of the interaction between the GS layer and the CG layer on the mechanical properties of the gradient materials remains to be studied. In this paper, single-sided GS Cu and double-sided GS Cu plates were prepared by surface mechanical attrition treatment (SMAT) on one side and both symmetrical sides respectively. It is found that GS samples have higher yield strength than CG samples, and the Cu samples with double-sided GS has an uniform elongation comparable to CG Cu. In addition, the global mutual constraint among different GS layers affect the mechanical behavior of gradient materials, which is revealed in the comparative study of double-sided and single-sided GS Cu. Specifically, the double-sided GS Cu exhibit higher yield strength and ductility than the single-sided GS Cu, because the former is better constrained. This study demonstrates that in addition to the local structural gradient, the symmetric constraint produce more effective HDI effect to enhance the mechanical properties. The global mutual constraints of GS layers played a positive role in producing superior mechanical properties.

Distributed information-theoretic clustering

Abstract We study a novel multi-terminal source coding setup motivated by the biclustering problem. Two separate encoders observe two i.i.d. sequences $X^n$ and $Y^n$, respectively. The goal is to find rate-limited encodings $f(x^n)$ and $g(z^n)$ that maximize the mutual information $\textrm{I}(\,{f(X^n)};{g(Y^n)})/n$. We discuss connections of this problem with hypothesis testing against independence, pattern recognition and the information bottleneck method. Improving previous cardinality bounds for the inner and outer bounds allows us to thoroughly study the special case of a binary symmetric source and to quantify the gap between the inner and the outer bound in this special case. Furthermore, we investigate a multiple description (MD) extension of the CEO problem with mutual information constraint. Surprisingly, this MD-CEO problem permits a tight single-letter characterization of the achievable region.

Perceiving more truth: A dilated-block-based convolutional network for rumor identification

To mitigate the negative influence of rumor spreading, rumor identification is urgently needed. Conventional feature-based identification models rely heavily on hand-crafted features and are thus not suitable for dynamic and complicated rumor identification scenarios. Although existing identification models based on deep neural networks eliminate manual effort and achieve automatic feature extraction, they can be further improved by modifying the underlying architecture—e.g., convolutional neural networks (CNNs). In this study, dilated convolution was used to form a dilated-block design architecture, with the aim of weakening the mutual constraint between the receptive field expansion and data loss in CNN-based models. In this architecture, a dilated convolution-based model called the “dilated-block-based convolutional network” (DBCN) is proposed. The DBCN stacks several dilated blocks to achieve a wider receptive field and automatically extract features from the input with less information loss. Accordingly, the constraint existing in CNN-based models are removed. An experiment on a rumor dataset showed that the DBCN model outperformed other baseline models in rumor identification,with an F1avg measure of 0.7719, which is an increase of 7.18% over the result of the previously best baseline model (the convolutional approach for misinformation identification). The DBCN model also performed well for early identification.

Supervisory Control of Petri Nets in the Presence of Replacement Attacks

This article addresses the robust control problem of discrete event systems assuming that replacement attacks may occur, thus making it appear that an event that has occurred looks like another event. In particular, we assume that this is done by tampering with the sensor-readings in the sensor communication channel. Specifically, we use Petri nets as the reference formalism to model the plant and assume a control specification in terms of a generalized mutual exclusion constraint. We propose three different methods to derive a control policy that is robust to the possible replacement attacks. The first two methods lead to an optimal (i.e., maximally permissive) policy but are computationally inefficient when applied to large-size systems. On the contrary, the third method computes a policy more efficiently and reveals more easily implementable in practice. However, this is done at the expense of optimality.

Framing a Phenomenological Mixed Method: From Inspiration to Guidance.

Despite a long history of researchers who combine phenomenology with qualitative or quantitative methods, there are only few examples of working with a phenomenological mixed method—a method where phenomenology informs both qualitative and quantitative data generation, analysis, and interpretation. Researchers have argued that in working with a phenomenological mixed method, there should be mutual constraint and enlightenment between the qualitative (first-person, subjective) and quantitative (third-person, objective) methods for studying consciousness. In this article, we discuss what a framework for phenomenological mixed methods could look like and we aim to provide guidance of how to work within such framework. We are inspired by resources coming from research in mixed methods and existing examples of phenomenological mixed-method research. We also present three cases of phenomenological mixed methods where we study complex social phenomena and discuss the process of how we conducted the studies. From both the research inspiration and our own studies, we depict the landscape of possibilities available for those interested in mixing phenomenology with qualitative and quantitative methods, as well as the challenges and common pitfalls that researchers face. To navigate in this landscape, we develop a three-fold structure, focusing on (1) the phenomenological frame, (2) the phenomenologically informed generation of qualitative and quantitative data (tier one), and (3) the phenomenologically informed analysis and interpretation of data (tier two).

Fixed-time position coordinated tracking control for spacecraft formation flying with collision avoidance

In this paper, the fixed-time stability of spacecraft formation reconfiguration (position tracking) is studied. Firstly, a novel nonsingular terminal sliding mode surface is designed and based on which a fixed-time coordinated controller is designed to keep the closed-loop system states have a finite settling time bounded by some predefined constants. Secondly, another nonsingular terminal sliding mode surface is designed by combining the artificial potential function and the aforementioned sliding surface, which meets the mutual distance constraint during transition process among spacecraft when it is bounded. Then another coordinated controller with fixed-time observer considering mutual distance constraint is presented, which guarantees the closed-loop system states stable also in bounded settling time. Finally, simulation results are shown to validate the correctness of the proposed theorems. It is worth mentioning that the control schemes also work even though there is a properly limit on the control input.

A Streamlined scRNA-Seq Data Analysis Framework Based on Improved Sparse Subspace Clustering

One advantage of single-cell RNA sequencing is its ability in revealing cell heterogeneity by cell clustering. However, cell clustering based on single-cell RNA sequencing is challenging due to the high transcript amplification noise, sparsity and outlier cell populations. In this study, we propose a novel sparse subspace clustering method called Structured Sparse Subspace Clustering and Completion for single-cell RNA sequencing analysis by assuming the cells related together are in the same subspace, and so the relationships among cells can be described within a subspace instead of between cell pairs. The proposed optimization model is solved by the Linearized Alternating Direction Method of Multipliers, in which data completion and spectral clustering are combined as a whole by mutual constraint. It is worth noting that random walk is used in the model to make the coefficient matrix more diagonal in the optimum iterative procedure, and the effect is significant. Our model is applied and compared with 5 state-of-the-art clustering methods on 6 public single cell datasets and a simulated data set with cell numbers varying from 56 to over 3000. As a result, our model outperforms the other clustering methods in clustering accuracy as evaluated by Adjusted Rand Index, Normalized Mutual Information, Homogeneity and Completeness, especially compared with the other improved sparse subspace clustering methods.

МЕХАНИЗМ ФУНКЦИОНИРОВАНИЯ ХУДОЖЕСТВЕННОГО ПРОЕКТИРОВАНИЯ В ДИЗАЙНЕ СРЕДЫ: МЕЖКОМПОНЕНТНЫЕ ОТНОШЕНИЯ ПРОТИВОРЕЧИЯ

The study aims to describe the structural relationships between the elements of the design process through understanding and describing the nature of mutual support and constraint. The research hypothesis assumes that achieving this goal will enable to explore the possibilities of influencing the process in order to regulate it. The study uses general scientific cognitive tools such as content analysis, synthesis, and the "wu-xing pentagram" symbolic method, which is a tool of categorical-systemic methodology. As a result, the mechanism of components communication: author — object — subject as the main in the design process is identified. Moreover, a model describing the structural relations between these components is developed and interpreted. The model allows to propose a mechanism for indirect control of the design process as a system activity. Finally, theoretical and practical results of further work are considered.

Contractualism and risk preferences

AbstractI evaluate two contractualist approaches to the ethics of risk: mutual constraint and the probabilistic, ex ante approach. After explaining how these approaches address problems in earlier interpretations of contractualism, I object that they fail to respond to diverse risk preferences in populations. Some people could reasonably reject the risk thresholds associated with these approaches. A strategy for addressing this objection is considering individual risk preferences, similar to those Buchak discusses concerning expected-utility approaches to risk. I defend the risk-preferences-adjusted (RISPREAD) contractualist approach, which calculates a population’s average risk preference and permits risk thresholds below that preference, only.

On the Variety and Veracity of Cyber Intrusion Alerts Synthesized by Generative Adversarial Networks

Many cyber attack actions can be observed, but the observables often exhibit intricate feature dependencies, non-homogeneity, and potentially rare yet critical samples. This work tests the ability to learn, model, and synthesize cyber intrusion alerts through Generative Adversarial Networks (GANs), which explore the feature space by reconciling between randomly generated samples and data that reflect a mixture of diverse attack behaviors without a priori knowledge. Through a comprehensive analysis using Jensen-Shannon Divergence, Conditional and Joint Entropy, and mode drops and additions, we show that the Wasserstein-GAN with Gradient Penalty and Mutual Information is more effective in learning to generate realistic alerts than models without Mutual Information constraints. We further show that the added Mutual Information constraint pushes the model to explore the feature space more thoroughly and increases the generation of low probability, yet critical, alert features. This research demonstrates the novel and promising application of unsupervised GANs to learn from limited yet diverse intrusion alerts to generate synthetic alerts that emulate critical dependencies, opening the door to proactive, data-driven cyber threat analyses.

Placing language in an integrated understanding system: Next steps toward human-level performance in neural language models

Language is crucial for human intelligence, but what exactly is its role? We take language to be a part of a system for understanding and communicating about situations. In humans, these abilities emerge gradually from experience and depend on domain-general principles of biological neural networks: connection-based learning, distributed representation, and context-sensitive, mutual constraint satisfaction-based processing. Current artificial language processing systems rely on the same domain general principles, embodied in artificial neural networks. Indeed, recent progress in this field depends on query-based attention, which extends the ability of these systems to exploit context and has contributed to remarkable breakthroughs. Nevertheless, most current models focus exclusively on language-internal tasks, limiting their ability to perform tasks that depend on understanding situations. These systems also lack memory for the contents of prior situations outside of a fixed contextual span. We describe the organization of the brain’s distributed understanding system, which includes a fast learning system that addresses the memory problem. We sketch a framework for future models of understanding drawing equally on cognitive neuroscience and artificial intelligence and exploiting query-based attention. We highlight relevant current directions and consider further developments needed to fully capture human-level language understanding in a computational system.

Analytical Solutions for Thermo-elastic Nonhomogeneous Equations

When the temperature changes, the stress produced by the external constraint and the mutual constraint between the internal parts of the object, which makes it unable to expand and contract freely, is called thermal stress. The thermal stress is caused by the steady temperature field. When the transient temperature changes to zero and the temperature distribution reaches a stable state, the temperature distribution can be obtained from the heat conduction equation and temperature boundary conditions, and then the displacement and stress can be obtained from the elastic equation including the temperature term.

SLIR: Synthesis, localization, inpainting, and registration for image-guided thermal ablation of liver tumors.

Thermal ablation is a minimally invasive procedure for treating small or unresectable tumors. Although CT is widely used for guiding ablation procedures, yet the contrast of tumors against normal soft tissues is often poor in CT scans, aggravating the accurate thermal ablation. In this paper, we propose a fast MR-CT image registration method to overlay pre-procedural MR (pMR) and pre-procedural CT (pCT) images onto an intra-procedural CT (iCT) image to guide the thermal ablation of liver tumors. At the pre-procedural stage, the Cycle-GAN model with mutual information constraint is employed to generate the synthesized CT (sCT) image from the input pMR. Then, pMR-pCT image registration is carried out via traditional mono-modal sCT-pCT image registration. At the intra-procedural stage, the region of the probe and its artifacts are automatically localized and inpainted in the iCT image. Then, an unsupervised registration network (UR-Net) is used to efficiently align the pCT with the inpainted iCT (inpCT) image. The final transform from pMR to iCT is obtained by concatenating the two estimated transforms, i.e., (i) from pMR image space to pCT image space (via sCT) and (ii) from pCT image space to iCT image space (via inpCT). The proposed method has been evaluated over a real clinical dataset and compared with state-of-the-art methods. Experimental results confirm that the proposed method achieves high registration accuracy with fast computation speed.

Dictionary learning with low mutual coherence constraint

This paper presents efficient algorithms for learning low-coherence dictionaries. First, a new algorithm based on proximal methods is proposed to solve the dictionary learning (DL) problem regularized with the mutual coherence of dictionary. This is unlike the previous approaches that solve a regularized problem where an approximate incoherence promoting term, instead of the mutual coherence, is used to encourage low-coherency. Then, a new solver is proposed for constrained low-coherence DL problem, i.e., a DL problem with an explicit constraint on the mutual coherence of the dictionary. As opposed to current methods, which follow a suboptimal two-step approach, the new algorithm directly solves the associated DL problem. Using previous studies, convergence of the new schemes to critical points of the associated cost functions is also provided. Furthermore, it is shown that the proposed algorithms have lower iteration complexity than existing algorithms. Our simulation results on learning low-coherence dictionaries for natural image patches as well as image classification based on discriminative over-complete dictionary learning demonstrate the superiority of the proposed algorithms compared with the state-of-the-art method.

Causal Beliefs and Perception of Temporal Order: The ‘Reordering Effect’

Empirical research on the relationship between temporality and causation is mostly dominated by the question of how temporal information constrains causal cognition. However, Bechlivanidis and Lagnado (2013, Psychol. Sci., 24, 1563–1572; 2016, Cognition, 146, 58–66) recently claimed to have discovered the ‘reordering effect’, in which causal beliefs have an influence on perception of temporal order. This paper argues for an attentional interpretation of this effect and suggests a solution to the circularity that arises from the mutual constraint between causal assumptions and perception of temporal order. Finally, it is shown how the reordering effect may challenge certain philosophical accounts of temporal illusions.

Heterogeneous microstructure-mediated ductile fracture of twin-roll cast Al–Mn strip

The microstructure and mechanical properties of the twin-roll cast (TRC) and traditional direct-chill cast (DC) Al–Mn strip have been compared using electron backscattered diffraction, scanning electron microscope, transmission electron microscope, and atom-probe tomography. The TRC Al–Mn strip showed a heterogeneous microstructure and texture along the thickness direction of the strip, forming a 'hard shell-soft core' structure. This heterogeneous microstructure resulted in a mutual constraint between the surface region and the center region, enabling the TRC strip to be plastically deformed to a total elongation >30% prior to fracture. In comparison, the DC strip, with a relatively homogeneous microstructure and texture, exhibited a uniform elongation close to zero. The distinct tensile properties presented by the two strips have been rationalized in terms of the differences in their microstructural features, including grain size/orientation, dislocation density, coarse constituent particles and solute atoms. A micromechanics model has been employed to understand the superior ductility of the TRC strip.

A novel zero watermark optimization algorithm based on Gabor transform and discrete cosine transform

SummaryZero watermark has no intervention on the carrier image, which is invisible in nature, and completely solves the problem of mutual constraint between robustness and invisibility in the traditional digital watermark technology. In order to improve the robustness of the zero watermark, a zero watermark algorithm based on Gabor transform and discrete cosine transform (DCT) is proposed in this paper. The algorithm performs Gabor transformation on the carrier image to obtain the directional characteristics of the image. Then it performs DCT and blocks, and performs singular value decomposition (SVD) on each subblock. The feature matrix is constructed using the largest eigenvalue of the subblock, and then XORed with the encrypted watermark to form a zero watermark. The algorithm utilizes the advantages of Gabor transform in the adjustment of scale and direction and selects the optimized parameters to extract image features. At the same time, the information concentration and decorrelation of DCT and SVD are utilized to improve the robustness and antiattack capability of the algorithm. The experimental results show that the proposed algorithm is robust to filtering, JPEG compression, noise, shearing, rotation, and other attacks, especially effective in antishear attack and rotation attack.

基于互结构正则约束的红外偏振图像增强算法

基于互结构正则约束的红外偏振图像增强算法

Computation of Admissible Marking Sets in Weighted Synchronization-Free Petri Nets by Dynamic Programming

We study the computation of admissible marking sets in generalized Petri nets. We first show that the admissibility checking in the generalized Petri net is NP-hard. Then, we consider a special subclass of generalized Petri nets called weighted-synchronization-free nets in which each transition has at most one input place. For a net in this subclass, we propose a generating function to compute by dynamic programming the set of admissible markings for a given generalized mutual exclusion constraint .

BiLSTM-CRF Chinese Named Entity Recognition Model with Attention Mechanism

In order to make up for the weakness of insufficient considering dependency of the input char sequence in the deep learning method of Chinese named entity recognition task, this paper proposes a method, which integrate Bidirectional Long Short-Term Memory (BiLSTM), attention mechanism and add the information of word vector. Firstly, the proposed model obtains the char vector feature extracted from the text corpus, which is then input to the BiLSTM model; Secondly, the attention mechanism is used to calculate the relevance between the current input char and the other input char of the BiLSTM model; Finally, the global feature is obtained according to relevance, concatenating the word vector feature, which is introduced to the Conditional random field(CRF) layer to perform the mutual constraint between tags. Thus, the classified result can be obtained. Based on the corpus of the Chinese Peoples’ Daily Newspaper in 1998, our experiments show that the proposed method can improved the performance and efficiency of named entity recognition, compared to the existing deep-learning method that combines word vector and char vector.