A bias–variance evaluation framework for information retrieval systems

In information retrieval (IR), the improvement of the effectiveness often sacrifices the stability of an IR system. To evaluate the stability, many risk-sensitive metrics have been proposed. Since the theoretical limitations, the current works study the effectiveness and stability separately, and have not explored the effectiveness–stability tradeoff. In this paper, we propose a Bias–Variance Tradeoff Evaluation (BV-Test) framework, based on the bias–variance decomposition of the mean squared error, to measure the overall performance (considering both effectiveness and stability) and the tradeoff between effectiveness and stability of a system. In this framework, we define generalized bias–variance metrics, based on the Cranfield-style experiment set-up where the document collection is fixed (across topics) or the set-up where document collection is a sample (per-topic). Compared with risk-sensitive evaluation methods, our work not only measures the effectiveness–stability tradeoff of a system, but also effectively tracks the source of system instability. Experiments on TREC Ad-hoc track (1993–1999) and Web track (2010–2014) show a clear effectiveness–stability tradeoff across topics and per-topic, and topic grouping and max–min normalization can effectively reduce the bias–variance tradeoff. Experimental results on TREC Session track (2010–2012) also show that the query reformulation and increase of user data are beneficial to both effectiveness and stability simultaneously.     

Finite-horizon H∞-consensus control for multi-agent systems with random parameters: The local condition case

This paper is concerned with the distributed H_∞-consensus control problem over the finite horizon for a class of discrete time-varying multi-agent systems with random parameters. First, by utilizing the proposed information matrix, a new formula is established to calculate the weighted covariance matrix of random matrix. Next, by allowing every agent to track the average of the neighbor agents, a novel local H_∞-consensus performance constraint is presented to cater to the local performance analysis. Then, by means of the proposed definition of the stochastic vector dissipativity-like over the finite horizon, a set of sufficient conditions for every agent is obtained such that the controlled outputs of the closed-loop multi-agent systems satisfy the proposed H_∞-consensus performance constraint. As a result, the proposed consensus control algorithm can be executed on each agent in an indeed distributed manner. Finally, a simulation example is employed to verify the effectiveness of the proposed algorithm.     

A new approach to H∞ filtering for linear time-delay systems

This paper proposes a class of $H_{\infty }$ filter design for continue-time systems with time-varying delay. Firstly, by exploiting a new Lyapunov function and using the convexity property of the matrix inequality, some delay-dependent stability conditions can be obtained for the asymptotical stability of the filtering-error system, which can lead to much less conservative analysis results. Secondly, based on the obtained conditions, the filter parameter matrixes can be obtained in terms of linear matrix inequalities (LMIs). Finally, two examples are given to demonstrate the effectiveness and the merit of the proposed method.     

A hybrid stochastic fractal search and local unimodal sampling based multistage PDF plus (1 + PI) controller for automatic generation control of power systems

This paper proposes to use a hybrid Stochastic Fractal Search (SFS) and Local Unimodal Sampling (LUS) based multistage Proportional Integral Derivative (PID) controller consisting of Proportional Derivative controller with derivative Filter (PDF) plus (1 + Proportional Integral) for Automatic Generation Control (AGC) of power systems. Initially, a single area multi-source power system consisting of thermal hydro and gas power plants is considered and parameters of Integral (I) controller is optimized by Stochastic Fractal Search (SFS) algorithm. The superiority of SFS algorithm over some recently proposed approaches such as optimal control, Differential Evolution (DE) and Teaching Learning Based Optimization (TLBO) is demonstrated. To improve the system performance further, LUS is subsequently employed. The study is further extended for different controllers like PID, and proposed multistage PID controller and the superiority of multistage PID controller over conventional PID controller structure is demonstrated. The study is further extended to a two-area six unit multi-source interconnected power system and the superiority of proposed approach over, TLBO and optimal control is demonstrated. Finally the study is extended to a three unequal area system power system with appropriate nonlinearities such as Generation Rate Constraint (GRC), Governor Dead Band (GDB) and time delay. From the analysis, it is found that hybrid SFS–LUS algorithm is superior to the original SFS algorithm and substantial improvement in system performance are realized with proposed multistage PID controller over conventional PID controller structure.     

A Sino-US center for eco-environment research inaugurated in Beijing

The signing ceremony for the framework agree- ment on the establishment of a China-US Joint Research Center for Ecosystem and Environ- mental Changes was held on July 20 on the campus of the CAS Institute of Geographic Sciences & Natural Resources Research (IGSNRR) in Beijing. Hosted by CAO Jie from the CAS Bureau of Inter- national Cooperation, the ceremony was attended by celebrities of the two sides in the field, including Direc- tor of the UT-ORNL Joint Institute for Biol…     

Stability in mean of partial variables for stochastic reaction–diffusion systems with Markovian switching

This paper is denoted to investigating stability in mean of partial variables for stochastic reaction–diffusion equations with Markovian switching (SRDEMS). By transforming the integral of the trajectory with respect to spatial variables as the solution of the stochastic ordinary differential equations with Markovian switching (SODEMS) and using Itô formula, sufficient criteria on uniform stability in mean, asymptotic stability in mean, uniformly asymptotic stability in mean, exponential stability in mean of partial variables for SRDEMS are first derived. An example is presented to illustrate the effectiveness and efficiency of the obtained results.     

SecureCPS: Cognitive inspired framework for detection of cyber attacks in cyber–physical systems

In the era of autonomous systems, the security is indispensable module for flexible computing environment. Due to increased computer power and network speed, a new computing paradigm, such as cognitive inspired computing, will emerge. Such a paradigm provides human-centered services that are convenient and enjoyable at any time, anywhere, and on any device. On the foundation of smart city environment, human computer interaction, intelligent services, and universal device connectivity, Cyber Physical Computing for Cyber Physical systems has recently been investigated. However, in this proposal, a cognitive inspired framework for securing CPS is scrutinized. The cognitive ability is conceded to the search engines by updating the PageRank ranking methodology. The proposed framework, named SecureCPS is trained with real time collective dataset for marking the relevancy of web page with the support the facial expressions. The eye regions are marked using Focal Point Detector algorithm. The framework is validated with machine learning models and resulted in achieving 98.51% accuracy and its outperforms the existing frameworks.     

A theoretical look at firm performance in high-tech organizations: what does existing theory tell us?

This paper looks at the changing nature of work for workers in knowledge-intensive high-tech firms. Four theories (institutional theory [IT], resource-based theory, game theory, and agency theory) are examined and propositions formulated as to how each theory helps explain the existence of intellectual cooperation and knowledge sharing within high-tech firms. The paper examines the usefulness of existing theory and suggests new research propositions, constructs, and methods to focus on building an empirical approach towards organizational research in the high-tech arena.     

Event-based weighted residual generator design via non-PDC scheme for fault diagnosis in T–S fuzzy systems

This paper is concerned with the event-based weighted residual generator design via non-parallel distribution compensation (PDC) scheme for fault diagnosis in discrete-time T–S fuzzy systems, under consideration of the imperfect premise matching membership functions. An event-triggered mechanism is firstly introduced to save communication resources, which leads to the premise variables of the system and observer to be asynchronous. Then, a fuzzy diagnostic observer with mismatched premise variables is designed to estimate the unmeasurable states of the system. Moreover, by using non-PDC method, a diagnostic observer-based weighted residual generator is established to improve the fault detection (FD) performance by using the information provided by each local system, in which the membership functions structure of the diagnostic observer and residual generator need not to be the same as the systems, and the L_∞/L₂ and L_∞ FD scheme is used to optimize the FD performance. Finally, two simulation results are provided to show the efficiency of the proposed non-PDC method.     

A new delay-range-dependent H∞ filter design for T–S nonlinear systems

This paper focuses on the problem of _H∞$H_{\infty }$ filter design for continuous-time Takagi and Sugeno (T–S) fuzzy systems with time-varying delays. The partitioning time delay technique is used to construct the Lyapunov–Krasovskii functional, furthermore, a novel delay-dependent _H∞$H_{\infty }$ filter design approach is proposed based on the matrix decoupling approach, and the filter parameters can be obtained by solving a set of linear matrix inequalities (LMIs). The numerical examples show that the proposed method is of less conservativeness than the existing methods.     

Modified uremic rat model — A tool for uremic studies in rats

Experimental model of chronic renal failure in rats have been described by many authors and has also been widely used in various studies. Many of these methods used highly sophisticated instruments which was difficult in our settings. The resection model was ideal in our experimental set-up, but this model had the risk of excessive bleeding and hypovolemia. In our study we used a combination of partial resection and ligation of the renal artery to create two models of stable uremia A and Moderate uremia B. Severe uremia. Both these models were compared with a sham operated group which served as controls. Following surgical procedure, the development of uremia was monitored by serial estimation of blood urea and serum creatinine levels that were measured at regular intervals (bi-weekly). From two weeks onwards the animals in the experimental group showed a significant elevation in the serum urea levels and a consistent elevation in the serum creatinine levels upto eight weeks when compared to the animals in the sham operated group. We established a modified method of producing renal failure which can be maintained for a period of six weeks. This model is simple, reproducible and less complicated that can be used for several studies relating to renal failure in the field of research.     

Disturbance observer-based fixed-time leader-following consensus control for multiple Euler–Lagrange systems: A non-singular terminal sliding mode scheme

This paper focuses on the fixed-time leader-following consensus problem for multiple Euler–Lagrange (EL) systems via non-singular terminal sliding mode control under a directed graph. Firstly, for each EL system, a local fixed-time disturbance observer is introduced to estimate the compound disturbance (including uncertain parameters and external disturbances) within a fixed time under the assumption that the disturbance is bounded. Next, a distributed fixed-time observer is designed to estimate the leader’s position and velocity, and the consensus problem is transformed into a local tracking problem by introducing such an observer. On the basis of the two types of observers designed, a novel non-singular terminal sliding surface is proposed to guarantee that the tracking errors on the sliding surface converge to zero within a fixed time. Furthermore, the presented control algorithm also ensures the fixed-time reachability of the sliding surface, while avoiding the singularity problem. Finally, the effectiveness of the proposed observers and control protocol is further verified by a numerical simulation.     

Safety in unpredictable complex systems – a framework for the analysis of safety derived from the nuclear power industry

The increasing complexity and power of our technologies compels us to find new ways in which to conceptualise, understand and maintain their safety in the long term. Some complex technological industries have performed better than others in terms of applying sustained and systematic approaches to the maintenance of safety. The United States nuclear power industry can be seen as an ideal test-bed for the development of safety initiatives, being responsible for the control of potentially unpredictable technology that involves extraordinary forces and costs. This paper describes and formalises a framework for better understanding the safety of complex socio-technological systems, based on key events in the development of safety in the United States nuclear power industry. The framework comprises two components: (1) a state-space approach for better conceptualising system failures, the benefits of incident reporting and remedial safety initiatives; and (2) a set of milestones that can be used to assess the development of safety in socio-technological industries. Healthcare and the United States nuclear power industry both represent complex socio-technological systems with similar technical characteristics. However, safety strategies in healthcare have not kept pace with the increasing complexity of clinical practice, and there have been international calls for improvements in patient safety. The framework is applied to the analysis of safety in healthcare, demonstrating its utility as an alternative safety analogy in healthcare. Use of the framework indicates substantial scope for improvements in healthcare safety through major evidence-based system redesign. By lowering the threshold for the reporting of incident data to include accident precursors, it is possible to identify problem areas before patient harm occurs.     

Science ＆ Technology for Sustainable Water in China A Perspective of the IAC Water Program and Studies in China

Water is a big issue in the world. As we enter the 21st century, a global water crisis threatens the security, stability and environmental sustainability of all nations, particularly those in the devel...     

Licensing standard-essential patents in the IoT – A value chain perspective on the markets for technology

Innovative industries need efficient markets for technology (MFTs). One determinant of MFT efficiency neglected until now is licensing level—that is, the level in the value chain where patents are licensed. Patents may be licensed upstream, to firms that put the patented knowledge into practice. I refer to this as integrated licensing. Alternatively, patents may be licensed further downstream in the value chain, in particular to makers of final devices. I call this bifurcated licensing since it separates intellectual property rights from the technical knowledge they cover. I study the licensing level of essential patents on communication standards such as LTE and Wi-Fi in relation to the Internet of Things (IoT). The choice of licensing level in this context is currently a hotly debated topic. To show how bifurcated licensing affects MFT efficiency, I present empirical evidence from a qualitative study comprising interviews with 30 individuals from 22 diverse firms, focusing on startups. IoT device makers clearly find the uncertainty regarding infringement, patent validity, and the licensing process hinders efficient licensing, which is compounded by the large number of IoT device makers and, for SMEs and startups, by resource constraints. As a theory contribution, I show that a patent's licensing level need not correspond with the implementation level of the patented knowledge—in other words, licensing may be bifurcated rather than integrated. I develop a model of how licensing level affects MFT efficiency. Implications for practice are that device-level licensing of standard-essential patents (SEPs), if broadly implemented, would have a negative effect on innovation and entrepreneurship in the IoT. Policymakers should ensure that SEP licensing is simplified.     

A taste for science? PhD scientists’ academic orientation and self-selection into research careers in industry

Recent research on industrial and academic science draws on the notion that academically trained scientists have a strong “taste for science”. However, little attention has been paid to potential heterogeneity in researchers’ taste for science and to potential selection effects into careers in industry versus academia. Using survey data from over 400 science and engineering PhD students, we examine the extent to which PhD students’ taste for science (e.g., desire for independence, publishing, peer recognition, and interest in basic research) and other individual characteristics predict preferences for research careers in industry versus academia. Our results suggest that PhD students who prefer industrial employment show a weaker “taste for science”, a greater concern for salary and access to resources, and a stronger interest in downstream work compared to PhD students who prefer an academic career. Our findings have important implications for innovation research as well as for managers and policy makers.     

Do policy mix characteristics matter for low-carbon innovation? A survey-based exploration of renewable power generation technologies in Germany

Policy mixes may play a crucial role in redirecting and accelerating innovation towards low-carbon solutions, thus addressing a key societal challenge. Towards this end, some argue that the characteristics of such policy mixes matter greatly, yet with little empirical evidence backing up such claims. In this paper we explore this link between policy mix characteristics and low-carbon innovation, using the research case of the transition of the German electricity system towards renewable energy. Our empirical insights are based on an innovation survey administered to German manufacturers of renewable power generation technologies which builds on the Community Innovation Survey. For our purposes we adjusted the survey to better capture companies’ perceptions of policy mixes. Employing a bivariate Tobit model we find evidence that companies’ perceptions regarding the consistency and credibility of the policy mix are positively associated with their innovation expenditures for renewable energies, and this positive link intensifies when considering the mutual interdependence of these policy mix characteristics. In contrast, neither the comprehensiveness of the instrument mix nor the coherence of policy processes were found to be related to innovation expenditures. Overall, these findings suggest that future research on low-carbon and eco-innovation should pay greater attention to the characteristics of policy mixes, rather than focusing on policy instruments only. Finally, our findings indicate a need to consider how policy may be measured in innovation surveys to generate better informed policy advice regarding the greening of innovation.