Incremental Strategy Research Articles

Multiobjective optimization by using cutting angle methods and hypervolume criterion

We translate a multiobjective optimization problem into a single objective Lipschitz problem by using the hypervolume criterion of the Pareto set. Deterministic global optimization methods allow one to track the whole Pareto front rather than converging to a single non-dominated solution. We augmented an efficient hypervolume computation technique with hypervolume increment strategy, and established Lipschitzianity of the resulting objective function. We used two deterministic Lipschitz optimization methods together with the hypervolume objective and benchmarked them against some state-of-the-art alternative multiobjective optimization methods, establishing the competitiveness of the proposed approach.

APM: Adaptive parameter multiplexing for class incremental learning

In recent developments within the domain of image classification, deep neural networks (DNNs) have attracted considerable scholarly interest and have been extensively trained using data in closed environments. Such training methodologies contrast sharply with the inherently open, progressive, and adaptive processes of the natural visual system, leading to emergent challenges. Among these, catastrophic forgetting is notable, where the network acquisition of new class information precipitates the erosion of previously established knowledge. Additionally, the network encounters the stability-plasticity dilemma, necessitating a delicate equilibrium between assimilating novel classes and retaining existing ones. To address these issues, we propose a novel incremental learning model, termed Adaptive Parameter Multiplexing (APM), which incorporates a cross-class parameter adaptive incremental strategy. Central to our methodology is the conceptualization of parameter multiplexing or incremental as a learnable optimization problem, enabling the model to autonomously evaluate and decide on the necessity for parameter adjustment throughout its training lifecycle. This framework is designed to enhance the ability of the network to extract features for new class categories effectively through incremental parameters while simultaneously employing parameter multiplexing to augment storage optimization. Our model is underpinned by a dual strategy of coarse-grained and fine-grained parameter multiplexing, guided by a learnable score that dynamically assesses the appropriateness of parameter multiplexing versus incremental updates, facilitating an optimized balance for incremental model performance and storage. In addition, we have integrated a novel regularization loss mechanism for the learnable score to optimize storage efficiency. The effectiveness of APM is empirically validated through rigorous testing on benchmark datasets, including ImageNet100, CIFAR100, CIFAR10, and CUB200. The experimental outcomes indicate that, with a trace amount of parameter increase, our model achieves significant enhancements in classification performance across both new and previously established classes, thereby surpassing existing benchmarks set by state-of-the-art algorithms in the field.

Інноваційні стратегії сталого розвитку підприємств в умовах мінливого ринку

The article analyzes various approaches to innovation development for enterprises operating in volatile markets. It explores critical innovation strategies, including proactive, reactive, imitative, niche, mergers and acquisitions (M&A), open innovation, incremental, and radical strategies. The proposed utilization of these strategies aims to ensure sustainable development for enterprises, considering their specific activities and market conditions. Proactive Strategies are characterized by their focus on anticipating and shaping market trends—proactive strategies position companies as market leaders by developing new products and services ahead of market demand. Analyzing technological trends and innovations is highlighted, enabling enterprises to maintain a competitive edge by staying ahead of competitors. Reactive strategies, however, adapt to changes in market and consumer demand. This approach involves responding swiftly to market shifts and competitive innovations. Imitative Strategies involve replicating or adapting innovations introduced by other companies. Niche Strategies concentrate on serving narrow market segments or specific technological niches. Companies utilizing niche strategies aim to become specialists in their chosen fields, offering unique solutions tailored to specific customer needs. This approach requires deep understanding, specialization, and expertise, allowing enterprises to command premium prices due to the distinctiveness of their products or services. This approach accelerates innovation processes by integrating external capabilities while reducing competitive pressures. However, successful implementation of M&A strategies demands meticulous planning and effective management to achieve synergy between organizations. In conclusion, the article offers a comprehensive framework for selecting and implementing innovative strategies tailored to enterprises’ specific needs and conditions in dynamic market environments. By strategically applying these approaches, enterprises can enhance their innovation capabilities, adapt to market changes effectively, and achieve sustainable development in the long term. Keywords: innovation strategies, sustainable development, dynamic market, proactive strategies, reactive strategies, imitative strategies, niche strategies, mergers and acquisitions, open innovation, incremental strategies, radical strategies, university partnerships, R&D, competitive advantage, market stability.

HeavySeparation: A Generic framework for stream processing faster and more accurate

Sketch as a probability data structure has been widely used in high volume, fast data streams. At the cost of a tiny accuracy in frequency estimation, it achieves a high speed with small memory usage. However, skewed data streams pose a significant challenge for existing sketches in terms of accuracy and speed using limited memory. To address this issue, we proposed a framework, called HeavySeparation, to enhance existing sketches by filtering elephant flow efficiently and accurately. We adopt a power-weakening increment strategy to allow sufficient competition in the early stages of identifying elephant flows and amplifying relative advantage when the frequency of candidate flow is large. To verify the effectiveness and efficiency of our framework, we apply the framework to two typical sketches and two common stream processing tasks. Results show that HeavySeparation framework reduces the error by around 1–2 orders of magnitude on average compared to the state-of-the-art in frequency estimation.

The economic and fiscal impact of incremental use of cell-based quadrivalent influenza vaccine for the prevention of seasonal influenza among healthcare workers in Italy

BackgroundSeasonal influenza has a significant impact on public health, generating substantial direct healthcare costs, production losses and fiscal effects. Understanding these consequences is crucial to effective decision-making and the development of preventive strategies. This study aimed to evaluate the economic and the fiscal impact of implementing an incremental strategy for seasonal influenza prevention using the cell-based quadrivalent influenza vaccine (QIVc) among healthcare workers (HCWs) in Italy.MethodsTo estimate the economic impact of implementing this strategy, we performed a cost analysis that considered direct healthcare costs, productivity losses and fiscal impact. The analysis considered a 3-year time horizon. A deterministic sensitivity analysis was also conducted.ResultsAssuming a vaccination coverage rate of 30% among HCWs, the analysis considered a total of 203 018 vaccinated subjects. On analysing the overall differential impact (including direct costs, indirect costs and fiscal impact), implementing QIVc vaccination as a preventative measure against influenza among HCWs in Italy would yield societal resource savings of €23 638.78 in the first year, €47 277.56 in the second year, and €70 916.35 in the third year, resulting in total resource savings of €141 832.69.ConclusionsThe study demonstrated that implementing the incremental use of QIVc as part of a preventive strategy for seasonal influenza among HCWs in Italy could yield positive economic outcomes, especially in terms of indirect costs and fiscal impact. The resources saved could be utilized to fund further public health interventions. Policy-makers should consider these findings when making decisions regarding influenza prevention strategies targeting HCWs.

Hybrid Scheduling of Tasks and Messages for TSN-Based Avionics Systems

Time-Sensitive Networking (TSN) has been widely considered as a promising networking candidate for avionics systems due to its capability of deterministic communication. In such TSN-based avionics systems, the network scheduling enables the timely transmission of messages. However, this is insufficient to satisfy the real-time requirements of functions, since functions involve the chain-execution of several tasks where messages only serve for the inter-task communication. In order to enhance the functionality of TSN-based avionics systems, scheduling should be extended from the network level to the system level to coordinate message transmission with task execution. Then, how to efficiently implement the hybrid scheduling of tasks and messages becomes an important issue. In this paper, we construct a novel hybrid scheduling framework for TSN-based avionics systems, which consists of system consistency constraints, in-loop function delay calculation, and two hybrid scheduling algorithms. Consistency constraints restrict the unexpected interaction of messages and tasks for hybrid scheduling to guarantee the system-level determinism. Function delay is the end-to-end delay of the task chain, and its calculation provides the optimization objective for hybrid scheduling indicated by two metrics, namely reaction delay and age delay. Scheduling algorithms improve solving efficiency and functional performance through the incremental strategies of message dynamic ordering and task-message rescheduling. Experimental results verify that, compared with exiting scheduling work which considers the dependency of messages on tasks, our work can complete complex scheduling for large systems even with hundreds of functions, and can reduce function delays by 69% in reaction delay and 37% in age delay.

An Accurate and Invertible Sketch for Super Spread Detection

Super spread detection has been widely applied in network management, recommender systems, and cyberspace security. It is more complicated than heavy hitter owing to the requirement of duplicate removal. Accurately detecting a super spread in real-time with small memory demands remains a nontrivial yet challenging issue. The previous work either had low accuracy or incurred heavy memory overhead and could not provide a precise cardinality estimation. This paper designed an invertible sketch for super spread detection with small memory demands and high accuracy. It introduces a power-weakening increment strategy that creates an environment encouraging sufficient competition at the early stages of discriminating a super spread and amplifying the comparative dominance to maintain accuracy. Extensive experiments have been performed based on actual Internet traffic traces and recommender system datasets. The trace-driven evaluation demonstrates that our sketch actualizes higher accuracy in super spread detection than state-of-the-art sketches. The super spread cardinality estimation error is 2.6–19.6 times lower than that of the previous algorithms.

SHED: A Newton-type algorithm for federated learning based on incremental Hessian eigenvector sharing

There is a growing interest in the distributed optimization framework that goes under the name of Federated Learning (FL). In particular, much attention is being turned to FL scenarios where the network is strongly heterogeneous in terms of communication resources (e.g., bandwidth) and data distribution. In these cases, communication between local machines (agents) and the central server (Master) is a main consideration. In this work, we present SHED, an original communication-constrained Newton-type (NT) algorithm designed to accelerate FL in such scenarios. SHED is by design robust to non independent identically distributed (non i.i.d.) data distributions, handles heterogeneity of agents’ communication resources (CRs), only requires sporadic Hessian computations, and achieves global asymptotic super-linear convergence. This is possible thanks to an incremental strategy, based on eigendecomposition of the local Hessian matrices, which exploits (possibly) outdated second-order information. SHED is thoroughly validated on real datasets by assessing (i) the number of communication rounds required for convergence, (ii) the overall amount of data transmitted and (iii) the number of local Hessian computations. For all these metrics, SHED shows superior performance against state-of-the art techniques like BFGS, GIANT and FedNL.

Artifact concreteness as imprinter in the organization design process

The fundamental purpose of design is to construct artifacts which interface inner and outer systems. Recent work in entrepreneurship makes a compelling argument that opportunities represent an artifact which interfaces inner systems of the entrepreneurial firm with outer systems of customers, markets and supplier stakeholders. In this formulation, opportunities might emerge from one of two ideal forms of design, namely experimentation or transformation. We build on such work, exploring a different artifact - the entrepreneurial organization. Adopting a design perspective, we compare the trajectories of two ventures with relatively similar starting points, but imprinted by different artifacts which encouraged the use of design approaches of experimentation in one and transformation in the other. We expose critical areas of divergence in the organizations that emerge. Imprinted by a trove of available concrete artifacts, the organization employing experimentation features dense, strongly tied networks, a more loyalty and work oriented culture and a simpler, less flexible, more incremental strategy. By contrast, the organization imprinted by penurious resources employed transformation, and features less dense, weakly tied networks, a more political, status oriented and discursive culture, and a more flexible, growth-oriented strategy. The practical implication of our work lies in the insight that through starting endowment imprint and design approach, future entrepreneurs might build features into the ultimate artifact they create - the organization itself.

A novel three-value grid scheme and rescue path planning algorithm for building fire

The 0-1 grid method is commonly used to divide a fire building into fully passable and fully impassable areas. Firefighters are only able to perform rescue tasks in the fully passable areas. However, in an actual building fire environment, there are three types of areas: fully impassable areas (areas blocked by obstacles or with heavy smoke and fire), fully passable areas, and partially passable areas (areas without obstacles or fire, but with some smoke risk). Due to the urgency of rescue, firefighters can consider conducting rescue tasks in both fully passable and partially passable areas to save valuable rescue time. To address this issue, we propose a three-value grid method, which classifies the fire environment into fully impassable areas, fully passable areas, and partially passable areas, represented by 1, 0, and 0.5, respectively. Considering that the ACO algorithm is prone to local optimum, we propose an enhanced ant colony algorithm (EACO) to solve the fire rescue path planning problem. The EACO introduces an adaptive heuristic function, a new pheromone increment strategy, and a pheromone segmentation rule to predict the shortest rescue path in the fire environment. Moreover, the EACO takes into account both the path length and the risk to balance rescue effectiveness and safety. Experiments show that the EACO obtains the shortest rescue path, which demonstrates its strong path planning capability. The three-value grid method and the path planning algorithm take reasonable application requirements into account.

CIPSO-Based Decision Support Method for Collision Avoidance of Super-Large Vessel in Port Waters

Effective and timely collision avoidance decision support is essential for super-large vessels navigating in port waters. To guarantee the navigational safety of super-large vessels, this work proposes a collision avoidance decision support method based on the curve increment strategy with adaptive particle swarm optimization (CIPSO). Firstly, the objective function is constructed based on the multi-objective optimization method. Here, a fuzzy comprehensive evaluation (FCE)-based vessel collision hazard model and vessel speed-varying energy-loss model integrating the Convention on the International Regulations for Preventing Collisions at Sea (COLREGS) are involved. Furthermore, in response to the limitations of the PSO algorithm, which is prone to falling into local optima in the later stages of iteration, a curve increment strategy is incorporated. To improve the performance of the global optimization, it is optimized using a local followed by global search method. The iterative evolution of CIPSO is used to obtain the optimal decision value in the set domain of feasible solutions. Finally, the effectiveness and feasibility of the proposed method are verified by the numerical simulation and large vessel maneuvering simulator, which can provide collision avoidance decision support for ship pilots.

SWINN: Efficient nearest neighbor search in sliding windows using graphs

Nearest neighbor search (NNS) is one of the main concerns in data stream applications since similarity queries can be used in multiple scenarios. Online NNS is usually performed on a sliding window by lazily scanning every element currently stored in the window. This paper proposes Sliding Window-based Incremental Nearest Neighbors (SWINN), a graph-based online search index algorithm for speeding up NNS in potentially never-ending and dynamic data stream tasks. Our proposal broadens the application of online NNS-based solutions, as even moderately large data buffers become impractical to handle when a naive NNS strategy is selected. SWINN enables efficient handling of large data buffers by using an incremental strategy to build and update a search graph supporting any distance metric. Vertices can be added and removed from the search graph. To keep the graph reliable for search queries, lightweight graph maintenance routines are run. According to experimental results, SWINN is significantly faster than performing a naive complete scan of the data buffer while keeping competitive search recall values. We also apply SWINN to online classification and regression tasks and show that our proposal is effective against popular online machine learning algorithms.

Mining Sequential Alarm Pattern Based on the Incremental Causality PrefixSpan Algorithm

Alarm flood is the main reason for the alarm system failure in the modern industry, so suppressing the subsequent alarms caused by anomalous propagation during the alarm flooding is an important issue. An incremental causality prefixSpan malgorithm is proposed to mine the frequent and sequence alarm patterns from the alarm flood, and dynamically update the old alarm mode from the new database. First, the alarm event is re-defined and the detection of alarm flood is formulated. The frequent alarm pattern with causality is expressed with the orderly and time delay features of sequential alarms. Under the new definition of alarm pattern, causality prefixSpan algorithm is proposed by introducing the time constraints, i.e., the causality. Furthermore, a new incremental mining strategy is co-implemented on the improved algorithm. When new alarm flooding data occurs, the old alarm modes are quickly updated without re-scanning the entire changing database. The incremental strategy makes the proposed method easily implemented in the real operation. The effectiveness of the proposed method is verified with the synthetic data sets and Tennessee-Sterman (TE) process.

Application of a Stochastic Gradient Descent-Based SVR Online Modeling Method to Time-Varying Forging Processes

In practice, forging processes have many unknown dynamics and fast time-varying characteristics due to transient load variation. This often results in insufficient data samples. Usually, support vector regression (SVR) can accurately model small sample data due to its good sparsity. However, using of sequential minimum optimization (SMO) algorithms increases the computational costs during the solution process, making it difficult to model fast time-varying systems. Aiming to address this problem, we developed an approach using online stochastic gradient descent (SGD) to improve SVR modeling efficiency. First, we used loss coefficients derived from the loss function to represent support vector loss. This effectively ensured sparsity in the modeling process. In this way, the SVR solving process was transformed into a loss coefficient calculation. This calculation was easy to achieve using SGD; thus, the solving process complexity was greatly reduced compared to SMO. On this basis, we developed an online incremental strategy to adapt the time-varying dynamics using online updating of step length, loss coefficients and bias term. Additional analysis demonstrated the convergence of the proposed online modeling method. Furthermore, modeling effect of this method is verified by using actual experiments with a 40MN isothermal die forging press.

In-Service Inspection & Repair strategy for newcleo Lead-cooled Fast Reactor

Newcleo is designing and building the next generation of nuclear lead-cooled fast reactors (LFRs) fueled with MOX. A fast-spectrum Small Modular Reactor of 30 MWe will be commissioned in 2030 (the Demonstrator) and a First-Of-Kind 200 MWe a few years later. To this end, newcleo develops an incremental strategy, starting first with the consolidation of mature and promising technologies with its own R&D laboratories and collaborations with major institutes and companies. This step aims to demonstrate the applicability of chosen technologies to a full mock-up of the reactor (the Precursor) available in 2026 for experiments. Inclusion of ISI&R (In Service Inspection and Repair) at the earliest stages of designs is a necessity to achieve a sufficient level of maturity and confidence in the technologies, and to increase safety, availability, and preservation of investments of experimental devices and reactors during their whole life.

Incremental adaptive fixed‐wing unmanned aerial vehicle maneuver control with nonlinear disturbance observer: A finite‐time approach

AbstractThis article proposes a versatile controller based on the incremental strategy to explore a finite‐time (FT) solution of fixed‐wing unmanned aerial vehicle (UAV) maneuver control considering the lumped disturbances such as unmodeled dynamics or exogenous perturbations. First, a finite‐time nonlinear disturbance observer augmented with a neural network item is developed to attenuate the adverse effect of the disturbances mentioned above. Then considering the estimated lumped disturbance, a finite‐time incremental controller in the quaternion form is proposed. Rigorous proof shows that the nonlinear disturbance observer and controller achieve finite‐time convergence. In the numerical simulation, the superiority of the proposed method compared to several advanced nonlinear control approaches is illustrated, despite the mismatches in aerodynamic coefficients, wind gusts, and actuator failures. Furthermore, based on a novel model identification approach, a hardware‐in‐the‐loop (HITL) experiment is performed in a high‐fidelity environment, wherein our method's feasibility and practicability are verified.

Offshore Hydrocarbon Exploitation Target Extraction Based on Time-Series Night Light Remote Sensing Images and Machine Learning Models: A Comparison of Six Machine Learning Algorithms and Their Multi-Feature Importance

The continuous acquisition of spatial distribution information for offshore hydrocarbon exploitation (OHE) targets is crucial for the research of marine carbon emission activities. The methodological framework based on time-series night light remote sensing images with a feature increment strategy coupled with machine learning models has become one of the most novel techniques for OHE target extraction in recent years. Its performance is mainly influenced by machine learning models, target features, and regional differences. However, there is still a lack of internal comparative studies on the different influencing factors in this framework. Therefore, based on this framework, we selected four different typical experimental regions within the hydrocarbon basins in the South China Sea to validate the extraction performance of six machine learning models (the classification and regression tree (CART), random forest (RF), artificial neural networks (ANN), support vector machine (SVM), Mahalanobis distance (MaD), and maximum likelihood classification (MLC)) using time-series VIIRS night light remote sensing images. On this basis, the influence of the regional differences and the importance of the multi-features were evaluated and analyzed. The results showed that (1) the RF model performed the best, with an average accuracy of 90.74%, which was much higher than the ANN, CART, SVM, MLC, and MaD. (2) The OHE targets with a lower light radiant intensity as well as a closer spatial location were the main subjects of the omission extraction, while the incorrect extractions were mostly caused by the intensive ship activities. (3) The coefficient of variation was the most important feature that affected the accuracy of the OHE target extraction, with a contribution rate of 26%. This was different from the commonly believed frequency feature in the existing research. In the context of global warming, this study can provide a valuable information reference for studies on OHE target extraction, carbon emission activity monitoring, and carbon emission dynamic assessment.

Sparsity promoting decentralized learning strategies for radio tomographic imaging using consensus based ADMM approach

Radio tomographic imaging (RTI) has wide applications in the detection and tracking of objects that do not require any sensor to be attached to the object. Consequently, it leads to device-free localization (DFL). RTI uses received signal strength (RSS) at different sensor nodes for imaging purposes. The attenuation maps, known as spatial loss fields (SLFs), measure the power loss at each pixel in the wireless sensor network (WSN) of interest. These SLFs help us to detect obstacles and aid in the imaging of objects. The centralized RTI system requires the information of all sensor nodes available at the fusion centre (FC), which in turn increases the communication overhead. Furthermore, the failure of links may lead to improper imaging in the RTI system. Hence, a distributed approach for the RTI system resolves such problems. In this paper, a consensus-based distributed strategy is used for distributed estimation of the SLF. The major contribution of this work is to propose a fully decentralized RTI system by using a consensus-based alternating direction method of multipliers (ADMM) algorithm to alleviate the practical issues with centralized and distributed incremental strategies. We proposed distributed consensus ADMM (DCADMM-RTI) and distributed sparse consensus ADMM (DSCADMM-RTI) for the RTI system to properly localize targets in a distributed fashion. Furthermore, the effect of quantization noise is verified by using the distributed consensus algorithms while sharing the quantized data among the neighbourhoods.

Mapping tourism strategy patterns on tourism area life cycle

PurposeTourist destinations are constantly changing products, evolving as per the controls exerted by the stakeholders. The study aims to map the pattern of tourism development and identify the strategies formed at the destination over a seven-decade period for a state as a unit of analysis.Design/methodology/approachThe paper evaluates tourism development through the tourism area life cycle (TALC) model and uses Mintzberg's strategy analysis process to identify strategies. The study involves time series analysis, pattern matching and explanation-building techniques. The TALC is plotted for the number of tourist arrivals from 1947 to 2019, and strategies are mapped for each stage.FindingsThe TALC shows a cycle-recycle pattern of tourism development. The research revealed several strategies at different stages. Both the central and state governments and entrepreneurs, distinctively and in conjunction, have formed strategies. The pattern shows the period of piecemeal and global strategic changes contributing to tourism development.Research limitations/implicationsThe research unearths the strategies that drive the development curves of TALC, emphasising the integration of TALC with other theories. The research also assesses the strategy formed in the pre-tourism stage.Practical implicationsThe research brings to light the use of TALC as a strategic road-mapping tool. In addition, the study emphasises the significance of global and piecemeal strategic periods and stakeholder's regulatory and operational roles.Originality/valueThe research uses a unique methodology that maps the strategies, periods of strategic changes and incremental strategies for each stage of TALC, along with identifying the stakeholders.

Loop Closure Detection via Locality Preserving Matching With Global Consensus

A critical component of visual simultaneous localization and mapping is loop closure detection (LCD), an operation judging whether a robot has come to a pre-visited area. Concretely, given a query image (i.e., the latest view observed by the robot), it proceeds by first exploring images with similar semantic information, followed by solving the relative relationship between candidate pairs in the 3D space. In this work, a novel appearance-based LCD system is proposed. Specifically, candidate frame selection is conducted via the combination of Super-features and aggregated selective match kernel (ASMK). We incorporate an incremental strategy into the vanilla ASMK to make it applied in the LCD task. It is demonstrated that this setting is memory-wise efficient and can achieve remarkable performance. To dig up consistent geometry between image pairs during loop closure verification, we propose a simple yet surprisingly effective feature matching algorithm, termed locality preserving matching with global consensus (LPM-GC). The major objective of LPM-GC is to retain the local neighborhood information of true feature correspondences between candidate pairs, where a global constraint is further designed to effectively remove false correspondences in challenging sceneries, e.g., containing numerous repetitive structures. Meanwhile, we derive a closed-form solution that enables our approach to provide reliable correspondences within only a few milliseconds. The performance of the proposed approach has been experimentally evaluated on ten publicly available and challenging datasets. Results show that our method can achieve better performance over the state-of-the-art in both feature matching and LCD tasks. We have released our code of LPM-GC at https://github.com/jiayi-ma/LPM-GC.