Iterative Scaling Research Articles

Multi-agent Gradient-Based Off-Policy Actor-Critic Algorithm for Distributed Reinforcement Learning

This paper proposes a gradient-based multi-agent actor-critic algorithm for off-policy reinforcement learning using importance sampling. Our algorithm is incremental with full gradients, and its complexity per iteration scales linearly with the size of approximation features. Previous multi-agent actor-critic algorithms are limited to the on-policy setting or off-policy emphatic temporal difference (TD) learning and they do not take advantage of the advances in off-policy gradient temporal difference learning (GTD). As a theoretical contribution, we establish that the critic step of the proposed algorithm converges to the TD solution of the projected Bellman equation and the actor step converges to the set of asymptotically stable fixed points. Numerical experiments on the multi-agent generalization of the Boyan’s chain problem show that the proposed approach provides improved performances in terms of stability and convergence rate as compared with the state-of-the-art baseline algorithm.

On some algorithms for estimation in Gaussian graphical models

Abstract In Gaussian graphical models, the likelihood equations must typically be solved iteratively. This paper investigates two algorithms: a version of iterative proportional scaling, which avoids inversion of large matrices, and an algorithm based on convex duality and operating on the covariance matrix by neighbourhood coordinate descent, which corresponds to the graphical lasso with zero penalty. For large, sparse graphs, the iterative proportional scaling algorithm appears feasible and has simple convergence properties. The algorithm based on neighbourhood coordinate descent is extremely fast and less dependent on sparsity, but needs a positive-definite starting value to converge. We provide an algorithm for finding such a starting value for graphs with low colouring number. As a consequence, we also obtain a simplified proof of existence of the maximum likelihood estimator in such cases.

Development and initial psychometric properties of the Research Complexity Index.

Research study complexity refers to variables that contribute to the difficulty of a clinical trial or study. This includes variables such as intervention type, design, sample, and data management. High complexity often requires more resources, advanced planning, and specialized expertise to execute studies effectively. However, there are limited instruments that scale study complexity across research designs. The purpose of this study was to develop and establish initial psychometric properties of an instrument that scales research study complexity. Technical and grammatical principles were followed to produce clear, concise items using language familiar to researchers. Items underwent face, content, and cognitive validity testing through quantitative surveys and qualitative interviews. Content validity indices were calculated, and iterative scale revision was performed. The instrument underwent pilot testing using 2 exemplar protocols, asking participants (n = 31) to score 25 items (e.g., study arms, data collection procedures). The instrument (Research Complexity Index) demonstrated face, content, and cognitive validity. Item mean and standard deviation ranged from 1.0 to 2.75 (Protocol 1) and 1.31 to 2.86 (Protocol 2). Corrected item-total correlations ranged from .030 to .618. Eight elements appear to be under correlated to other elements. Cronbach's alpha was 0.586 (Protocol 1) and 0.764 (Protocol 2). Inter-rater reliability was fair (kappa = 0.338). Initial pilot testing demonstrates face, content, and cognitive validity, moderate internal consistency reliability and fair inter-rater reliability. Further refinement of the instrument may increase reliability thus providing a comprehensive method to assess study complexity and related resource quantification (e.g., staffing requirements).

Optimization of clothing style and color design mode and innovation of technique under symbiosis mode

Abstract In this paper, we combine the thesaurus matching through the clothing style feature mining method, get the conditional probability index through great likelihood estimation, use iterative scale training parameters, and calculate the similarity mean of the belonging feature values. To identify and classify clothing styles, texture features are extracted from clothing styles, and the optimal training sample set of popular clothing colors is selected. The highest data mining algorithm in symbiotic mode has an accuracy rate of 96.4% for short-skirt clothing styles, a statistical result of 82.61 for tie-dye clothing techniques, and a maximum color threshold of 0.14. Symbiotic mode achieves a visual effect of distinctive features, which leads to optimal clothing styles and color innovations.

Developing a Theory of Change for a Digital Youth Mental Health Service (Moderated Online Social Therapy): Mixed Methods Knowledge Synthesis Study.

Common challenges in the youth mental health system include low access, poor uptake, poor adherence, and limited overall effectiveness. Digital technologies offer promise, yet challenges in real-world integration and uptake persist. Moderated Online Social Therapy (MOST) aims to overcome these problems by integrating a comprehensive digital platform into existing youth mental health services. Theory of change (ToC) frameworks can help articulate how and why complex interventions work and what conditions are required for success. The objective of this study is to create a ToC for MOST to explain how it works, why it works, who benefits and how, and what conditions are required for its success. We used a multimethod approach to construct a ToC for MOST. The synthesis aimed to assess the real-world impact of MOST, a digital platform designed to enhance face-to-face youth mental health services, and to guide its iterative refinement. Data were gathered from 2 completed and 4 ongoing randomized controlled trials, 11 pilot studies, and over 1000 co-design sessions using MOST. Additionally, published qualitative findings from diverse clinical contexts and a review of related digital mental health literature were included. The study culminated in an updated ToC framework informed by expert feedback. The final ToC was produced in both narrative and table form and captured components common in program logic and ToC frameworks. The MOST ToC captured several assumptions about digital mental health adoption, including factors such as the readiness of young people and service providers to embrace digital platforms. External considerations included high service demand and a potential lack of infrastructure to support integration. Young people and service providers face several challenges and pain points MOST seeks to address, such as limited accessibility, high demand, poor engagement, and a lack of personalized support. Self-determination theory, transdiagnostic psychological treatment approaches, and evidence-based implementation theories and their associated mechanisms are drawn upon to frame the intervention components that make up the platform. Platform usage data are captured and linked to short-, medium-, and long-term intended outcomes, such as reductions in mental health symptoms, improvements in functioning and quality of life, reductions in hospital visits, and reduced overall mental health care costs. The MOST ToC serves as a strategic framework for refining MOST over time. The creation of the ToC helped guide the development of therapeutic content personalization, user engagement enhancement, and clinician adoption through specialized implementation frameworks. While powerful, the ToC approach has its limitations, such as a lack of standardized methodology and the amount of resourcing required for its development. Nonetheless, it provides an invaluable roadmap for iterative development, evaluation, and scaling of MOST and offers a replicable model for other digital health interventions aiming for targeted, evidence-based impact.

Classical iterative proportional scaling of log-linear models with rational maximum likelihood estimator

In this work we investigate multipartition models, the subset of log-linear models for which one can perform the classical iterative proportional scaling (IPS) algorithm to numerically compute the maximum likelihood estimate (MLE). Multipartition models include families of models such as hierarchical models and balanced, stratified staged trees. We define a sufficient condition, called the Generalized Running Intersection Property (GRIP), on the matrix representation of a multipartition model under which the classical IPS algorithm produces the exact MLE in one cycle. In this case, the MLE is a rational function of the data. Additionally we connect the GRIP to the toric fiber product and to previous results for hierarchical models and balanced, stratified staged trees. This leads to a characterization of balanced, stratified staged trees in terms of the GRIP.

An Effective Data-driven Audio Watermarking Technique

Abstract: Audio piracy remains a persistent issue, highlighting the crucial need to protect audio signals and respect intellectual property rights. To tackle this challenge, we propose a two-step procedure consisting of intrinsic time scale decomposition (ITD) and Singular Value Decomposition (SVD).In our approach, the audio to be watermarked is initially divided into intrinsic mode functions and a low-frequency residue component using an iterative time scale decomposition technique. Subsequently, we apply Singular Value Decomposition to both the low-frequency component and the watermark. This allows us to embed the watermark within the residue component of the audio.To generate the watermarked audio, we combine the intrinsic mode functions with the watermarked residue component using inverse Intrinsic Time Scale Decomposition. This ensures that the watermark becomes integrated into the audio signal. Notably, our proposed approach effectively addresses computational complexity, a common challenge, while also prioritizing robustness and imperceptibility in line with advanced methods employed in the field.

Pulse wave signal modelling and feature extraction based on Lognormal function from photoplethysmography in wireless body area networks

Among the various physiological signals of the human body, the pulse wave signal from photoplethysmography is widely used for commercial products in Wireless Body Area Networks. The comfortable wearing of a product and the richness of physiological information are two advantages of Wireless Body Area Networks. To address the shortcomings of the existing modelling pulse wave signal in lacking characteristics that are consistent with a human physiological mechanism in medicine, we proposed a Lognormal function model of the pulse wave signal and a novel model to extract physiological parameters. The proposed Lognormal function model consists of four successively spaced single-peaked pulses with long trailing features. To define the parameters in the proposed Lognormal function model, we proposed a method that took advantage of the positivity and negativity of the first-order derivatives and the trans-zero point of the second-order derivatives. Based on the determined parameters, we introduced a four-shot staged curve fitting approach that can displace the sum of the four fits at iterative and different time scales. Finally, a parameter vector with 12 elements, which is known as a physiological feature to determine the health status of the human body in Wireless Body Area Networks. Experimental results show that the proposed Lognormal function model is superior to the conventional Gaussian function model in terms of physiological importance and waveform fitting accuracy.

Dimensions of nonbinary antiprimitive BCH codes and some conjectures

Bose-Chaudhuri-Hocquenghem (BCH) codes have been intensively investigated. Even so, there is only a little known about primitive BCH codes, let alone non-primitive ones. In this paper, let q>2 be a prime power, the dimension of a family of non-primitive BCH codes of length n=qm+1 (also called antiprimitive) is studied. These codes are also linear codes with complementary duals (called LCD codes). Through some approaches such as iterative algorithm, partition and scaling, all coset leaders of Cx modulo n with q⌈m2⌉<x≤2q⌈m2⌉+2 are given for m≥4. And for odd m the first several largest coset leaders modulo n are determined. Furthermore, a new kind of sequences is introduced to determine the second largest coset leader modulo n with m even and q odd. Also, for even m some conjectures about the first several coset leaders modulo n are proposed, whose complete verification would wipe out the difficult problem to determine the first several coset leaders of antiprimitive BCH codes. After deriving the cardinalities of the coset leaders, we shall calculate exact dimensions of many antiprimitive LCD BCH codes.

Scale Validation of the Mandarin-Language Supportive Care Needs Survey–Adolescent Form

To (1) modify the Mandarin-language 34-item Supportive Care Needs Survey-Adult Form into the Adolescent Form and (2) examine the psychometric properties of the Adolescent Form. A multiphase, iterative scale validation process was used in this methodological study. Participants who were 13 to 18 years old and receiving cancer treatment in inpatient or outpatient settings, or receiving follow-up care in outpatient settings, were recruited using a convenience sampling method. Confirmatory factor analysis demonstrated good fitness of indices, and all factor loadings for the 18-item Adolescent Form were >0.50, which supported the scale's construct validity. The Adolescent Form score was significantly correlated with the symptom distress score (r = 0.56, P < .01) and quality of life score (r = -0.65, P < .01), which indicated the scale's convergent validity. The correlated item-total correlations (0.30-0.78), Cronbach's alpha (.93), and test-retest reliability coefficient (0.79) confirmed the scale's stability. This study successfully modified the 34-item Adult Form into the 18-item Adolescent Form. Given its adequate psychometric properties, this concise scale has great promise as a useful, feasible, and age-appropriate tool for evaluating care needs among adolescents with cancer who speak Mandarin. This scale can screen for unmet care needs in busy pediatric oncology settings or large-scale clinical trials. It allows for cross-sectional comparison of unmet care needs between adolescent and adult populations and for longitudinal follow-up into how unmet care needs change from adolescence into adulthood.

An Iterative Scale Purification Procedure on l z for the Detection of Aberrant Responses

Many person-fit statistics have been proposed to detect aberrant response behaviors (e.g., cheating, guessing). Among them, l z is one of the most widely used indices. The computation of l z assumes the item and person parameters are known. In reality, they often have to be estimated from data. The better the estimation, the better l z will perform. When aberrant behaviors occur, the person and item parameter estimations are inaccurate, which in turn degrade the performance of l z. In this study, an iterative procedure was developed to attain more accurate person parameter estimates for improved performance of l z. A series of simulations were conducted to evaluate the iterative procedure under two conditions of item parameters, known and unknown, and three aberrant response styles of difficulty-sharing cheating, random-sharing cheating, and random guessing. The results demonstrated the superiority of the iterative procedure over the non-iterative one in maintaining control of Type-I error rates and improving the power of detecting aberrant responses. The proposed procedure was applied to a high-stake intelligence test.

Toric geometry of entropic regularization

Entropic regularization is a method for large-scale linear programming. Geometrically, one traces intersections of the feasible polytope with scaled toric varieties, starting at the Birch point. We compare this to log-barrier methods, with reciprocal linear spaces, starting at the analytic center. We revisit entropic regularization for unbalanced optimal transport, and we develop the use of optimal conic couplings. We compute the degree of the associated toric variety, and we explore algorithms like iterative scaling.

Rasch development and validation of the "Medellín Dry Eye Inventory" (ME·Dry), an instrument to measure multidimensional quality of life compromise in patients with dry eye disease.

Dry eye disease (DED) is a common condition that affects the quality of life. There is a great need for better-developed scales that comply with Rasch model requirements. Prospective study including patients with DED. A series of focus groups were performed to determine the best items to be included. A Rasch modeling methodology was used to validate the Medellín Dry Eye Inventory (ME·Dry). After iterative analysis and scale modification, a final version of the scale was attained which complied with the Rasch analysis expectations. Correlation between the different subscales of the ME·Dry and the Ocular Surface Disease Index (OSDI) was evaluated through Spearman correlation. A total of 166 patients with DED were included. Rasch modeling demonstrated an excellent behavior for the ME·Dry, including four subscales: Symptoms, Triggers, Activity Limitation, and Emotional Compromise. Infit and Outfit parameters were all between 0.50 and 1.50, with excellent category utilization. Person and item separation and reliability were excellent for all subscales. There was a need for a category collapsing for the Emotional Compromise subscale. There was a strong correlation between the different subscales of the ME·Dry except for the Emotional Compromise subscale, which seems to be independent. The ME·Dry is a reliable scale, complying with the Rasch model expectations, that allows for a reliable measurement of quality of life compromise in patients with DED. Emotional compromise secondary to DED does not seem to correlate with disease severity as assessed by the other quality-of-life subscales.

Integreted strategic and tactical optimization planning of biomass to bioethanol supply chains coupled with operational plan using vehicle routing: A case study in Ethiopia

The optimization of supply chain, a subject of active research in process system engineering, frequently leads to large Mixed Integer Linear or Non-Linear Problems more particularly when applied to realistic industrial cases. In this domain, one main challenge concerns multi scale modeling to accurately integrate the different decision scales (from strategic to operational) and to study the impact of the decision across these scales, in order to effectively contributes to intelligent and coherent decision making. Unfortunately, very few papers cover the three decision levels - strategic (long-term decisions: supply network design), tactical (medium-term decisions: management of network design) and operational (short- term decisions: routing, inventory management…) to have a holistic perspective of the supply chain. Therefore, the first objective of this paper is to model as finely as possible the three decision levels by focusing more particularly on the transport related decision due to their importance on the supply chain performances. As a progress, the Vehicle Routing Problem, one of the most studied problem in logistic and one of the most complex, is integrated into a strategic and tactical model. But, the resulting large-scale MILP model must be solved in a reasonable computational time. To achieve this second objective, we developed a novel iterative multi scale strategy that optimizes the model at two spatial and two temporal scales. This strategy is based on a deep knowledge of the mathematical models and especially of the variables that strongly influence the resolution. The real biofuel supply chain in the Ethiopian country is used to demonstrate the capabilities of the approach, and the improvement in computing time and solution strategy. In addition to the change in the design of the supply chain network, the use of an operational model that finely describes the transport activities shows that the costs related to these activities are often underestimated by a factor between 2 and 3 in the classical models of the literature. The main perspectives of this work are to address the two main shortcomings of the proposal: writing and solving in a reasonable computational time a single model rather than using a decomposition (especially useful for multi-objective optimization) and integrating other decisions at the operational level to not be limited to those related to transportation.

DDCAL: Evenly Distributing Data into Low Variance Clusters Based on Iterative Feature Scaling

This work studies the problem of clustering one-dimensional data points such that they are evenly distributed over a given number of low variance clusters. One application is the visualization of data on choropleth maps or on business process models, but without over-emphasizing outliers. This enables the detection and differentiation of smaller clusters. The problem is tackled based on a heuristic algorithm called DDCAL (1d distribution cluster algorithm) that is based on iterative feature scaling which generates stable results of clusters. The effectiveness of the DDCAL algorithm is shown based on 5 artificial data sets with different distributions and 4 real-world data sets reflecting different use cases. Moreover, the results from DDCAL, by using these data sets, are compared to 11 existing clustering algorithms. The application of the DDCAL algorithm is illustrated through the visualization of pandemic and population data on choropleth maps as well as process mining results on process models.

Motivating Proactive Biorisk Management.

Scholars and practitioners of biosafety and biosecurity (collectively, biorisk management or BRM) have argued that life scientists should play a more proactive role in monitoring their work for potential risks, mitigating harm, and seeking help as necessary. However, most efforts to promote proactive BRM have focused on training life scientists in technical skills and have largely ignored the extent to which life scientists wish to use them (ie, their motivation). In this article, we argue that efforts to promote proactive BRM would benefit from a greater focus on life scientists' motivation. We review relevant literature on life scientists' motivation to practice BRM, offer examples of successful interventions from adjacent fields, and outline ideas for possible interventions to promote proactive BRM, along with strategies for iterative development, testing, and scaling.

Evaluation of the quality of measurements of the testing laboratory during interlaboratory comparative tests

The article presents the statistical processing of the test results of the mass fraction of the chlorine ion (Cl) obtained from two accredited testing laboratories. The stages of statistical processing of the results of interlaboratory comparison tests are as follows: elimination of gross errors, determination of the uniformity of the sample, calculation of the assigned value, evaluation of the quality of the measurement results. The presence of gross errors in the sample was determined by the Grubbs test, and the uniformity of the sample was confirmed by the Fisher’s test and Student’s test. The assigned value was calculated by a robust method using algorithm A with an iterative scale. The quality of laboratory measurement results was evaluated using the z-score. The z-score value for each participant showed that the results of measuring the mass fraction of chlorine ion in cement samples were acceptable. The quality of the results of these laboratories is satisfactory and does not require corrective measures. These results of the test quality assessment can be used for the organizers of interlaboratory tests, as well as for accredited laboratories that check their qualifications.

Distributed Multi-hop Clustering Approach with Low Energy Consumption in WSN

The purpose of sensing the environment and geographical positions, device monitoring, and information gathering are accomplished using Wireless Sensor Network (WSN), which is a non-dependent device consisting of a distinct collection of Sensor Node (SN). Thus, a clustering based on Energy Efficient (EE), one of the most crucial processes performed in WSN with distinct environments, is utilized. In order to efficiently manage energy allocation during sensing and communication, the present research on managing energy efficiency is performed on the basis of distributed algorithm. Multiples of EE methods were incapable of supporting EE routing with MIN-EC in WSN in spite of the focus of EE methods on energy harvesting and minimum Energy Consumption (EC). The three stages of performance are proposed in this research work. At the outset, during routing and Route Searching Time (RST) with fluctuating node density and PKTs, EC is reduced by the Hybrid Energy-based Multi-User Routing (HEMUR) model proposed in this work. Energy efficiency and an ideal route for various SNs with distinct PKTs in WSN are obtained by this model. By utilizing the Approximation Algorithm (AA), the Bregman Tensor Approximation Clustering (BTAC) is applied to improve the Route Path Selection (RPS) efficiency for Data Packet Transmission (DPT) at the Sink Node (SkN). The enhanced Network Throughput Rate (NTR) and low DPT Delay are provided by BTAC. To MAX the Clustering Efficiency (CE) and minimize the EC, the Energy Effective Distributed Multi-hop Clustering (GISEDC) method based on Generalized Iterative Scaling is implemented. The Multi-User Routing (MUR) is used by the HEMUR model to enhance the EC by 20% during routing. When compared with other advanced techniques, the Average Energy Per Packet (AEPP) is enhanced by 39% with the application of proportional fairness with Boltzmann Distribution (BD). The Gaussian Fast Linear Combinations (GFLC) with AA are applied by BTAC method with an enhanced Communication Overhead (COH) for an increase in performance by 19% and minimize the DPT delay by 23%. When compared with the rest of the advanced techniques, CE is enhanced by 8% and EC by 27% with the application of GISEDC method.

Decoding Nonbinary LDPC Codes via Proximal-ADMM Approach

In this paper, we focus on decoding nonbinary low-density parity-check (LDPC) codes in Galois fields of characteristic two via the proximal alternating direction method of multipliers (proximal-ADMM). By exploiting Flanagan/Constant-Weighting embedding techniques and the decomposition technique based on three-variables parity-check equations, two efficient proximal-ADMM decoders for nonbinary LDPC codes are proposed. We show that both of them are theoretically guaranteed convergent to some stationary point of the decoding model and either of their computational complexities in each proximal-ADMM iteration scales linearly with LDPC code’s length and the size of the considered Galois field. Moreover, the decoder based on the Constant-Weight embedding technique satisfies the favorable property of codeword symmetry. Simulation results demonstrate their effectiveness in comparison with state-of-the-art LDPC decoders.

General Adaptation Scale for International Students: Development and Validation

Embracing a holistic and eclectic perspective, this study aims to develop and empirically test an instrument to assess international students’ general adaptation to a host country. The data were obtained from a sample of international undergraduate students (n=843) studying at a Turkish state university using purposive sampling methods of maximum variation. The analyses were conducted in two consecutive phases with two different groups of international students. Upon a meticulous data clean-up and preliminary analyses for the assumptions of normality and reliability, exploratory factor analysis (EFA) to discover factor structure was utilized. In the second phase, confirmatory factor analysis (CFA) was performed through Lisrel to validate the scale structure revealed by EFA. After a rigorous and iterative scale development process, the results confirm the reliability of factors, model fit and construct validity of the General Adaptation Scale for International Students (GASIS). GASIS as a multi-dimensional instrument consists of 28 items using a five-point Likert scale with four factors labelled as academic adaptation, sociocultural adaptation, psychological adaptation, and daily life adaptation to assess international students’general adaptation levels. The final form of 28-item GASIS with psychometric features, as well as implications and limitations for future research are included in the study.