Adaptive Iteration Research Articles

An Iterative Resampling Deep Decoupling Domain Adaptation method for class-imbalance bearing fault diagnosis under variant working conditions

Bearing fault diagnosis suffers from unsatisfied accuracy under variant working conditions, due to class-imbalance and distribution discrepancy of faulty data. Class-imbalance will cause classification bias that makes the diagnostic model tend to misclassify faulty classes, while the distribution discrepancy caused by variant working conditions will further reduce the model’s generalization. Thus, this paper is mainly focusing on solving the two challenges simultaneously, and an Iterative Resampling Deep Decoupling Domain Adaptation (IRDDDA) model is proposed. The model consists of a feature extractor, domain classifier, label classifier and feature resampler modules, and it adopts a decoupling two-stage training framework. In the first stage, the model learns domain invariant features from the source domain and target domain based on original class-imbalance datasets. In the second stage, the feature extractor is partially fixed, and the feature-based resampling method is used to adjust the biased classifier caused by the class-imbalance problem. An iterative training strategy is adopted to simultaneously update the upper-layers of the feature extractor and adjust the classifier till convergence. The method is verified on the CWRU open dataset and the dataset collected from our bearing test rig. The verification results show that our method can achieve a better fault diagnostic performance overall, especially in minority faulty classes.

Multi-scale iterative domain adaptation for specific emitter identification

Multi-scale iterative domain adaptation for specific emitter identification

BREATHE-T1D: Using iterative mixed methods to adapt a mindfulness-based intervention for adolescents with type 1 diabetes: Design and development

BackgroundNegative affect is prevalent among adolescents with type 1 diabetes (T1D) and may impact diabetes self-management and outcomes through stress-related behaviors such as disordered eating. MethodsWe describe the development of and design for the adaptation of a mindfulness-based intervention (MBI) for adolescents with T1D and negative affect. BREATHE-T1D is an MBI designed to target negative affect that has been tailored to address the unique lived experiences of adolescents with T1D. Qualitative interviews with stakeholders and participants were used to inform iterative adaptations to the intervention and control curricula over the course of the study. The primary aim of this paper is to describe the design, development, and protocol of the present pilot feasibility trial. ConclusionsIterative, qualitative methodology throughout the adaptation of an intervention is important for ensuring the resulting intervention is relevant and meaningful for the target population.Clinical Trial Registration Number: NCT05268393

Development of an intervention to support parents receiving treatment in psychiatric inpatient hospital using participatory design methods.

When parents of dependent children are treated in psychiatric inpatient hospital, it typically involves separation of parent and child for the duration of treatment, which can be highly distressing to the dyad and can result in disruption to the parent-child relationship. Parents who have experienced hospitalisation have expressed a desire for their parenting identity to be recognized and appropriately engaged with during their treatment. This recognition includes provision of interventions which support them as parents to limit the impact of their mental health on their children. The current study, the first of its kind known to have taken place, details a collaborative intervention development project for parents receiving inpatient care. The current study, the first of its kind known to have taken place, details a collaborative intervention development project for parents receiving inpatient care. This project involved the adaptation and extension of a prior parenting-focused course for parents high in anxiety to meet the needs of parents being treated in inpatient settings. In the first two stages of the three-phase project, patients, carers and mental health practitioners contributed to the revision and delivery plan for the course including developing new content for the intervention. In the final stage, which took the form of a participatory evaluation, the intervention was delivered to 11 parents receiving inpatient treatment who then provided extensive feedback. A series of iterative adaptations to the intervention were made in response to this feedback alongside stakeholder input. The final intervention comprises five modules focused on exploring the experience of parents alongside specific learning and skills orientated toward boosting their connection with their children during hospitalisation and in readiness for discharge. Preliminary feedback from patients and ward staff has been positive and the process of delivering the project on inpatient wards was associated with no increase in negative clinical outcomes. The successful development of a targeted intervention within inpatient psychiatric units offers a signal that parents treated in this setting welcome the opportunity to be supported in their parenting role. As the first known UK intervention of its kind to be developed in partnership with patients, ward staff and management, it is specifically tailored to the context and needs of this group with the potential to be delivered by a range of health professionals in this setting.

A Reconstruction Algorithm for Compression Perception Based on Local Learning Dictionary and Sparsity Iterative Update

The global sparse representation method based on compressed sensing fails to capture the local texture and detail structure of an image. To address this, a local dictionary learning method based on the wavelet domain is proposed. The wavelet high-frequency subband sub-block classification and local dictionary learning are implemented using the FCM clustering method and K-L method, respectively. At the reconstruction end, a compressed sampling Matching pursuit algorithm based on iterative update of sparsity is proposed. This adaptive iteration can effectively reconstruct the original image under unknown sparsity conditions. Simulation experiments show that compared to existing reconstruction methods, this approach has the advantages of simple computation, strong adaptive sparse representation ability for the original image, and superior reconstructed image performance.

Surrogate optimisation strategies for intraocular lens formula constant optimisation.

To investigate surrogate optimisation (SO) as a modern, purely data-driven, nonlinear adaptive iterative strategy for lens formula constant optimisation in intraocular lens power calculation. A SO algorithm was implemented for optimising the root mean squared formula prediction error (rmsPE, defined as predicted refraction minus achieved refraction) for the SRKT, Hoffer Q, Holladay, Haigis and Castrop formulae in a dataset of N = 888 cataractous eyes with implantation of the Hoya Vivinex hydrophobic acrylic aspheric lens. A Gaussian Process estimator was used as the model, and the SO was initialised with equidistant datapoints within box constraints, and the number of iterations restricted to either 200 (SRKT, Hoffer Q, Holladay) or 700 (Haigis, Castrop). The performance of the algorithm was compared to the classical gradient-based Levenberg-Marquardt algorithm. The SO algorithm showed stable convergence after fewer than 50/150 iterations (SRKT, HofferQ, Holladay, Haigis, Castrop). The rmsPE was reduced systematically to 0.4407/0.4288/0.4265/0.3711/0.3449 dioptres. The final constants were A = 119.2709, pACD = 5.7359, SF = 1.9688, -a0 = 0.5914/a1 = 0.3570/a2 = 0.1970, C = 0.3171/H = 0.2053/R = 0.0947 for the SRKT, Hoffer Q, Holladay, Haigis and Castrop formula and matched the respective constants optimised in previous studies. The SO proves to be a powerful adaptive nonlinear iteration algorithm for formula constant optimisation, even in formulae with one or more constants. It acts independently of a gradient and is in general able to search within a (box) constrained parameter space for the best solution, even where there are multiple local minima of the target function.

Examining the dynamics of social cohesion: A call for a different perspective on scaling impacts of real-world laboratories

Social cohesion is an important impact category for scaling real-world laboratory experiments. This idea has been largely overlooked in the transformative research debate. Based on observations within real-world laboratories that focused on iterative, co-creative, and practice-based climate change adaptation, we identify social cohesion, first, as a prerequisite for real-world laboratory impacts. Second, social cohesion can itself be an impact, enhancing the scaling potential of real-world laboratories. Cooperation can pave the way for amplifying real-world laboratories’ activities temporally and spatially.

From programming-to-modeling-to-prompts smart ubiquitous applications

Since its introduction by Mark Weiser, ubiquitous computing has received increased interest in the dawn of technological advancement. Supported by wireless technology advancement, embedded systems, miniaturization, and the integration of various intelligent and communicative devise, context-aware ubiquitous applications actively and intelligently use rich contextual information to assist their users. However, their designs are subject to continuous changes imposed by external factors. Nowadays, software engineering, particularly in the fields of Model-Driven Engineering, displays a strong tendency towards developing applications for pervasive computing. This trend is also fueled by the rise of generative artificial intelligence, paving the way for a new generation of no-code development tools and models specifically trained on open-source code repositories to generate applications from their descriptions. The specificities of our approach lies in starting with a graphical model expressed using a domain-specific language (DSL) composed of symbols and formal notations. This allows for graphically instantiating and editing applications, guiding and assisting experts from various engineering fields in defining ubiquitous applications that are eventually transformed into peculiar models. We believe that creating intelligent models is the best way to promote software development efficiency. We have used and evaluated recurrent neural networks, leveraging the recurrence of processing the same contextual information collected within this model, and enabling iterative adaptation to future evolutions in ubiquitous systems. We propose a prototype instantiated by our meta-model which tracks the movements of individuals who were positive for COVID-19 and confirmed to be contagious. Different deep learning models and classical machine learning techniques are considered and compared for the task of detection/classification of COVID-19. Results obtained from all techniques were evaluated with confusion matrices, accuracy, precision, recall and F1-score. In summary, most of the results are very impressive. Our deep learning approach used a RNN architecture produced up to 92.1% accuracy. With the recent development of OpenAI Codex, optimized for programming languages, we provided the same requirements to the Codex model and asked it to generate the source code for the COVID-19 application, comparing it with the application generated by our workshop.

Adaptive measurement strategy for quantum subspace methods

Estimation of physical observables for unknown quantum states is an important problem that underlies a wide range of fields, including quantum information processing, quantum physics, and quantum chemistry. In the context of quantum computation, in particular, existing studies have mainly focused on holistic state tomography or estimation on specific observables with known classical descriptions, while this lacks the important class of problems where the estimation target itself relies on the measurement outcome. In this work, we propose an adaptive measurement optimization method that is useful for the quantum subspace methods, namely the variational simulation methods that utilize classical postprocessing on measurement outcomes. The proposed method first determines the measurement protocol for classically simulatable states, and then adaptively updates the protocol of quantum subspace expansion (QSE) according to the quantum measurement result. As a numerical demonstration, we have shown for excited-state simulation of molecules that (i) we are able to reduce the number of measurements by an order of magnitude by constructing an appropriate measurement strategy (ii) the adaptive iteration converges successfully even for a strongly correlated molecule of H4. Our work reveals that the potential of the QSE method can be empowered by elaborated measurement protocols, and opens a path to further pursue efficient quantum measurement techniques in practical computations.

Coping Power-Rural: Iterative Adaptation of an Evidence-Based Preventive Intervention for Rural Upper Elementary and Middle Schools

BackgroundEducators in rural schools are uniquely situated to address youth mental health disparities, yet often face challenges in delivering mental health supports. This paper describes the process of adapting the evidence-based Coping Power program, a small group prevention program for youth with aggressive behavior problems, to be a two-tiered (Tier 1 and Tier 2), transdiagnostic intervention to improve fit and feasibility for rural upper elementary and middle schools.MethodIdentified challenges with the Coping Power program for rural areas included program length, substantial staffing and resource requirements, lack of universal programming, low caregiver engagement, and co-occurring problems. Initial adaptations included a classroom and small group format implemented by school staff, teacher consultations integrated into coaching and co-facilitation, and a technology-supported caregiver component. Implementer feedback forms, coaching notes, and individual interviews informed the iterative development and feasibility testing process.ResultsBetween 2019 and 2023, thirteen schools across six rural districts implemented the program. Student curriculum revisions included order and relative emphasis of content, classroom and small group overlap, necessary simplification of concepts, improved contextualization to the rural setting, and the addition of student workbooks. Supports for implementers included fully developed lesson plans and slides, a comprehensive implementation manual, video lesson overviews, action-focused training, and a 3-session coaching model to support implementer preparation and sustain motivation. Teacher and caregiver infographic text “nudges” were improved to promote generalization of concepts across settings.DiscussionBy partnering with school-based implementers, the adapted program holds promise to be more feasible and appealing for rural schools than the original model. This fully developed program is now ready for larger-scale testing in rural schools.

An Improved Adaptive Iterative Extended Kalman Filter Based on Variational Bayesian

The presence of unknown heavy-tailed noise can lead to inaccuracies in measurements and processes, resulting in instability in nonlinear systems. Various estimation methods for heavy-tailed noise exist. However, these methods often trade estimation accuracy for algorithm complexity and parameter sensitivity. To tackle this challenge, we introduced an improved variational Bayesian (VB)-based adaptive iterative extended Kalman filter. In this VB framework, the inverse Wishart distributionis used as the prior for the state prediction covariance matrix. The system state and noise parameter posterior distributions are then iteratively updated for adaptive estimation. Furthermore, we make adaptive adjustments to the IEKF filter parameters to enhance sensitivity and filtering accuracy, thus ensuring robust prediction estimation. A two-dimensional target tracking and nonlinear numerical UNGM simulation validated our algorithm. Compared to existing algorithms RKF-ML and GA-VB, our method showed significant improvements in RMSEpos and RMSEvel, with increases of 21.81% and 22.11% respectively, and a 49.04% faster convergence speed. These results highlight the method’s reliability and adaptability.

A unique numerical iterative approach for modelling individual phase stress-strain curves in dual phase steel

Understanding the effects of martensite volume fractions (V m) in dual phase (DP) steel resulting from heat treatment is crucial for designing structures for mechanical impact resistance and optimizing manufacturing processes. DP steel’s material behaviour depends heavily on its microstructure properties. While stress-strain curves for individual phases in DP steels are often determined using empirical models, extensive experimental data is required to establish empirical model constants. This research aims to achieve two main objectives: firstly, to calibrate stress-strain curves for pure ferrite and pure martensite using limited experimental data using micromechanical adaptive iteration algorithm (MAIA). This calibration involves using stress-strain data from DP steels with varying V m during the calibration stage and additional data for verification. Secondly, to conduct a comprehensive sensitivity analysis of MAIA to assess its capabilities and limitations. Microstructure-based finite element (FE) models, simulated with ABAQUS/Standard, are employed to predict stress-strain curves under uniaxial tensile test conditions. The MAIA approach successfully calculated ferrite and martensite stress-strain curves that could predict plastic behaviour of DP steel with different V m, which agreed with experimental work. Key advantages of this approach include avoiding complex 3D microstructure geometries and requiring only two experimentally obtained stress-strain curves with different Vm for material constant calibration, along with another curve for validation. However, the experimental data selected for calibration must have a V m difference between 20%–50% and one of the DP steels must have a low martensite volume fraction. The FE micromechanical model could capture the effect of softening of martensite phase and strengthening of ferrite phase as compared to its bulk properties for DP steel. The effect of V m on strain hardening rate was also successfully captured. This technique comes with obvious shortcomings, such as excluding crystal plasticity behaviour, and change in chemical composition within the individual phase with varying martensite volume fraction.

The defect feature extraction of ultrasonic phased array detection based on adaptive region growth.

An ultrasonic phased array defect extraction method based on adaptive region growth is proposed, aiming at problems such as difficulty in defect identification and extraction caused by noise interference and complex structure of the detected object during ultrasonic phased array detection. First, bilateral filtering and grayscale processing techniques are employed for the purpose of noise reduction and initial data processing. Following this, the maximum sound pressure within the designated focusing region serves as the seed point. An adaptive region iteration method is subsequently employed to execute automatic threshold capture and region growth. In addition, mathematical morphology is applied to extract the processed defect features. In the final stage, two sets of B-scan images depicting hole defects of varying sizes are utilized for experimental validation of the proposed algorithm's effectiveness and applicability. The defect features extracted through this algorithm are then compared and analyzed alongside the histogram threshold method, Otsu method, K-means clustering algorithm, and a modified iterative method. The results reveal that the margin of error between the measured results and the actual defect sizes is less than 13%, representing a significant enhancement in the precision of defect feature extraction. Consequently, this method establishes a dependable foundation of data for subsequent tasks, such as defect localization and quantitative and qualitative analysis.

ITERATIVE ADAPTATION ALGORITHMS IN MULTICRITERIA TASKS

This article examines using of iterative adaptation algorithms to solve the problem of determining measurement location of the carotid artery intima-media complex. The formulation of a multi-criteria decision-making problem, as the basis for determining correct criterion for proper selection and successful recognizing of the required object in an ultrasound image. The work discusses principles of constructing cascade classifiers as well as, the use of the cascade Haar classifier and the cascade LBP classifier, for which Haar primitives and local binary templates are used as a basis. The results of experimental studies in order to determine effectivity of different boosting algorithms to solve this problem are presented. The best results were shown by the Haar cascade classifier, developed using an iterative adaptation algorithm, which manages solving a multicriteria problem on a given training set more successfully and determines the most suitable areas for measuring the thickness of the common carotid artery intimate media complex.

Retrieval of Tree Height Percentiles over Rugged Mountain Areas via Target Response Waveform of Satellite Lidar

The retrieval of tree height percentiles from satellite lidar waveforms observed over mountainous areas is greatly challenging due to the broadening and overlapping of the ground return and vegetation return. To accurately represent the shape distributions of the vegetation and ground returns, the target response waveform (TRW) is resolved using a Richardson–Lucy deconvolution algorithm with adaptive iteration. Meanwhile, the ground return is identified as the TRW component within a 4.6 m ground signal extent above the end point of the TRW. Based on the cumulative TRW distribution, the height metrics of the energy percentiles of 25%, 50%, 75%, and 95% are determined using their vertical distances relative to the ground elevation in this study. To validate the proposed algorithm, we select the received waveforms of the Global Ecosystem Dynamics Investigation (GEDI) lidar over the Pahvant Mountains of central Utah, USA. The results reveal that the resolved TRWs closely resemble the actual target response waveforms from the coincident airborne lidar data, with the mean values of the coefficient of correlation, total bias, and root-mean-square error (RMSE) taking values of 0.92, 0.0813, and 0.0016, respectively. In addition, the accuracies of the derived height percentiles from the proposed algorithm are greatly improved compared with the conventional Gaussian decomposition method and the slope-adaptive waveform metrics method. The mean bias and RMSE values decrease by the mean values of 1.68 m and 2.32 m and 1.96 m and 2.72 m, respectively. This demonstrates that the proposed algorithm can eliminate the broadening and overlapping of the ground return and vegetation return and presents good potential in the extraction of forest structure parameters over rugged mountainous areas.

Adaptive double-loop coverage optimization of underwater wireless directional restricted sensor networks

Underwater Wireless Sensor Networks (UWSNs) play a crucial role in underwater communication system, in which node deployment is a complex problem. Most of the previous researches are based on all-direction underwater acoustic sensors. However, most of the sensors are in limited direction in practical systems. The deployment of directional sensors not only needs to consider its position adjustment, but also needs to accurately adjust its direction according to the sensing angle, sensing distance and other parameters in its sensing model, so it is difficult to achieve optimization with traditional algorithms. In this paper, we propose the Adaptive Heuristic Algorithm for UWSNs (AHAU) to solve this problem in underwater communication systems. AHAU features the addition of inner layer of adaptive parameter iteration to outer layer of global iteration. AHAU can more precisely adjust the direction of the sensor to achieve higher target coverage rate. The effectiveness and reliability of the proposed algorithm are verified by designing experiments. The results of experiments show that AHAU can achieve better adjustment effect under different parameter conditions to ensure the spatial coverage of UWSNs.

Iterative evaluation of mobile computer-assisted digital chest x-ray screening for TB improves efficiency, yield, and outcomes in Nigeria.

Wellness on Wheels (WoW) is a model of mobile systematic tuberculosis (TB) screening of high-risk populations combining digital chest radiography with computer-aided automated detection (CAD) and chronic cough screening to identify presumptive TB clients in communities, health facilities, and prisons in Nigeria. The model evolves to address technical, political, and sustainability challenges. Screening methods were iteratively refined to balance TB yield and feasibility across heterogeneous populations. Performance metrics were compared over time. Screening volumes, risk mix, number needed to screen (NNS), number needed to test (NNT), sample loss, TB treatment initiation and outcomes. Efforts to mitigate losses along the diagnostic cascade were tracked. Persons with high CAD4TB score (≥80), who tested negative on a single spot GeneXpert were followed-up to assess TB status at six months. An experimental calibration method achieved a viable CAD threshold for testing. High risk groups and key stakeholders were engaged. Operations evolved in real time to fix problems. Incremental improvements in mean client volumes (128 to 140/day), target group inclusion (92% to 93%), on-site testing (84% to 86%), TB treatment initiation (87% to 91%), and TB treatment success (71% to 85%) were recorded. Attention to those as highest risk boosted efficiency (the NNT declined from 8.2 ± SD8.2 to 7.6 ± SD7.7). Clinical diagnosis was added after follow-up among those with ≥ 80 CAD scores and initially spot -sputum negative found 11 additional TB cases (6.3%) after 121 person-years of follow-up. Iterative adaptation in response to performance metrics foster feasible, acceptable, and efficient TB case-finding in Nigeria. High CAD scores can identify subclinical TB and those at risk of progression to bacteriologically-confirmed TB disease in the near term.

Variable step size VLF/ELF nonlinear channel adaptive filtering algorithm based on Sigmoid function

The signals received by very low-frequency/extremely low-frequency nonlinear receivers are frequently affected by intense atmospheric pulse noise stemming from thunderstorms and global lightning activity. Current noise processing algorithms designed for nonlinear channels within these frequency ranges, which are predicated on fractional p-order moment alpha stable distribution criteria (where 0 < p < α < 2, and p and α denote distinct characteristic indices of alpha stable distribution noise), are constrained by their reliance on limited p-order moment statistics. As a result, the performance of low-frequency nonlinear channel receivers experiences significant degradation when confronted with robust pulse noise interference (0 < p < α < 2). To tackle this challenge, the present study introduces a novel variable step robust mixed norm (RMN) adaptive filtering algorithm, designated as SVS-RMN, which is based on the Sigmoid function. Leveraging the nonlinearity of the Sigmoid function and building upon the power function Hammerstein nonlinear channel model, the algorithm aims to enhance the RMN algorithm by deriving new cost functions and adaptive iteration formulas. The performance of the proposed algorithm is evaluated in comparison to conventional RMN algorithms based on fractional low-order moment (FLOM) criteria (0 < p < 2), as well as other algorithms employing variable step sizes and either FLOM or radial basis function (RBF) criteria, across various intensities of pulse noise and mixed signal-to-noise ratios. The experimental results reveal the following: (1) The proposed algorithm effectively mitigates strong pulse noise interference and significantly enhances the tracking performance of the RMN algorithm compared to conventional RMN algorithms based on FLOM criteria. (2) In terms of computational efficiency, simplicity of structure, convergence speed, and stability, the proposed algorithm surpasses other algorithms based on FLOM or RBF criteria.

Renewing Iterative Self-Labeling Domain Adaptation With Application to the Spine Motion Prediction

The area of transfer learning comprises supervised machine learning methods that cope with the issue when the training and testing data have different input feature spaces or distributions. In this work, we propose a novel transfer learning algorithm called Renewing Iterative Self-labeling Domain Adaptation (Re-ISDA). This work is motivated by a cervical spine motion prediction problem, with the goal of predicting the cervical spine motion of different subjects using the measurements of “exterior features”. The joint distribution of the exterior features for each subject may vary with one’s distinct BMI, sex and other characteristics; the sample size of this problem is limited due to the high experimental cost. The proposed method is well suited for transfer learning problems with limited training samples. The learning problem is formulated as a dynamic programming model and the latter is then solved by an efficient greedy algorithm. Numerical studies show that the proposed method outperforms prevailing transfer learning methods. The proposed method also achieves high prediction accuracy for the cervical spine motion problem. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This paper is motivated by the problem of using human body exterior features (e.g., contours, surface markers) acquired by computer vision or other measurement means to determine the underlying physics or mechanics. The goal is to build a subject-specific model to predict cervical spine motion based on surface-based measurements. Conventional supervised learning and transfer learning methods fail to give an accurate prediction due to the subject-to-subject variation and/or the small sample size of the training data. This paper proposes a novel transfer learning method called renewing iterative self-labeling domain adaptation to modify the possible error of the pseudo labels and thus make the training process more stable. The proposed method improves the accuracy by over <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$35\%$</tex-math> </inline-formula> on the traditional transfer learning methods. This method is applicable to limited-sample problems with different training and testing input distributions, and is particularly suitable for time-series data.

Evaluation and lessons learned from the dissemination and implementation science scholars program in the national cancer prevention and control research network

Abstract Background: The Centers for Disease Control and Prevention (CDC)-funded Cancer Prevention and Control Research Network (CPCRN) has been a leader in cancer-related dissemination &amp; implementation (D&amp;I) science. Given increased demand for D&amp;I research, the CPCRN Scholars Program launched in 2021 to expand the number of practitioners, researchers, and trainees proficient in cancer D&amp;I science methods. Methods: The evaluation was informed by a logic model and data collected through electronic surveys. Through an application process (baseline survey), we assessed scholars’ competencies in D&amp;I science domains/subdomains, collected demographic data, and asked scholars to share proposed project ideas. We distributed an exit survey one month after program completion to assess scholars’ experience and engagement with the program and changes in D&amp;I competencies. A follow-up survey was administered to alumni nine months post-program to measure their continued network engagement, accomplishments, and skills. Results: Three cohorts completed the program, consisting of 20, 17, and 25 scholars in Years 1-3, respectively. There was a significant increase in the total D&amp;I competency scores for all three cohorts for 4 overarching domains and 43 subdomains (MPre = 1.38 MPost = 1.89). Differences were greatest for the domain of Practice-Based Considerations (0.50 mean difference) and Theory &amp; Analysis (0.47 mean difference). Alumni surveys revealed that scholars appreciated access to D&amp;I-focused webinars, toolkits, and training resources. 80% remain engaged with CPCRN workgroups and investigators. Conclusions: Program evaluation with scholars and alumni helped with ongoing quality assurance, introspection, and iterative program adaptation to meet scholars’ needs. This approach is recommended for large-scale capacity-building training programs.