National Research Project Research Articles

Leonhard Med, a trusted research environment for processing sensitive research data.

This paper provides an overview of the development and operation of the Leonhard Med Trusted Research Environment (TRE) at ETH Zurich. Leonhard Med gives scientific researchers the ability to securely work on sensitive research data. We give an overview of the user perspective, the legal framework for processing sensitive data, design history, current status, and operations. Leonhard Med is an efficient, highly secure Trusted Research Environment for data processing, hosted at ETH Zurich and operated by the Scientific IT Services (SIS) of ETH. It provides a full stack of security controls that allow researchers to store, access, manage, and process sensitive data according to Swiss legislation and ETH Zurich Data Protection policies. In addition, Leonhard Med fulfills the BioMedIT Information Security Policies and is compatible with international data protection laws and therefore can be utilized within the scope of national and international collaboration research projects. Initially designed as a "bare-metal" High-Performance Computing (HPC) platform to achieve maximum performance, Leonhard Med was later re-designed as a virtualized, private cloud platform to offer more flexibility to its customers. Sensitive data can be analyzed in secure, segregated spaces called tenants. Technical and Organizational Measures (TOMs) are in place to assure the confidentiality, integrity, and availability of sensitive data. At the same time, Leonhard Med ensures broad access to cutting-edge research software, especially for the analysis of human -omics data and other personalized health applications.

Research on electromagnetic pulse signal detection method based on intelligent compressed sensing technology

Abstract Electromagnetic pulse signal measurement is crucial for national security, scientific research, technological advancement, and emergency preparedness, playing a vital role in various aspects of modern society. In this work, an electromagnetic pulse signal detection approach utilizing compressed sensing technology is suggested. By unitizing three parallel low-frequency analog-to-digital converters, the high-frequency acquisition waveform of the measured electromagnetic pulse signal can be recovered. The compressed sensing effect of the proposed system on electromagnetic pulse signal is tested by simulation, which verifies the feasibility of this method.

Analysis of the National Research and Innovation Program for the years 2024–2027

The article analysed the National Research and Innovation Program for the years 2024–2027 from the point of view of its compliance with policy documents and its relevance for the development of research in the Republic of Moldova. Elements of good quality of the program (especially identified problems and risks) and less successful ones are characterized. The latter highlighted the poor correlation between existing problems and proposed solutions, indicators that are not relevant for measuring the evolution of science and the general and formal character of most of the planned actions. The analysis showed that the program will not be able to play an important role in changing the situation in the field and in the development of the national scientific research system and most likely the science policy decision-making process will continue to be parallel to the provisions of the policy documents

Does OER Research-Based Learning Improve Performance: A Case Study from Students Enrolled in a Community College at City University of New York (CUNY)

This study investigates the effectiveness of Open Education Resources (OER) and research-based learning. A quantitative analysis was conducted on students enrolled in Macroeconomic courses at a City University of New York (CUNY) community college in Spring 2018 and Spring 2019. Prior studies have shown the positive effects of conducting research in natural sciences. Yet, there is a lot to be inquired about regarding the impact of research-based learning on social sciences and the use of OER materials as a cheaper and more equitable alternative to commercial textbooks. Additionally, this study evaluates the academic effectiveness of research-based learning through assessment scores through a new indicator, the growth mindset profile. Findings should be used to assess critically research-based learning applications incorporated with OER materials and open pedagogy, as well as to consider the duration, wide application throughout the program rather than isolated courses, and design of course assignments and research projects to engage students actively and effectively.

Risk-Averse Markov Decision Processes Through a Distributional Lens

By adopting a distributional viewpoint on law-invariant convex risk measures, we construct dynamic risk measures (DRMs) at the distributional level. We then apply these DRMs to investigate Markov decision processes, incorporating latent costs, random actions, and weakly continuous transition kernels. Furthermore, the proposed DRMs allow risk aversion to change dynamically. Under mild assumptions, we derive a dynamic programming principle and show the existence of an optimal policy in both finite and infinite time horizons. Moreover, we provide a sufficient condition for the optimality of deterministic actions. For illustration, we conclude the paper with examples from optimal liquidation with limit order books and autonomous driving. Funding: This work was supported by Natural Sciences and Engineering Research Council of Canada [Grants RGPAS-2018-522715 and RGPIN-2018-05705].

Living labs in agrifood studies: An opportunity to revisit fundamental questions about participatory research?

The use of the term ‘living lab’ is now widespread, and in agrifood studies has been incorporated into research funding requirements. However, the methodological skills and prerequisites for successful living labs are not particularly well articulated. Based on our experience (five national and EU research projects) and our research backgrounds in (interdisciplinary) social science and participatory research, we highlight several risks associated with the use of living labs in agrifood studies without having the resources, reflectivity or skills required. These risks include: 1) imposing predefined agendas that are masked as participatory processes, 2) placing an overemphasis on confidentiality or knowledge sharing, and 3) getting lost in researcher roles. We argue that there is a need to discuss and improve understanding of these risks in order to produce successful living labs in agrifood studies, as well as avoid conceptual dilution or confusion.

Finnish Pediatric Surgery Hub – From Centralization to Collective Learning and Sharing of Expertise

Aim of the studyContinuous surgical developments, growing awareness of patient representatives and limited health-care resources are pushing for innovative approaches to ensure equal high-quality pediatric surgical care. We aimed to describe early experiences and assess surgical safety of a novel nationwide pediatric surgery collaborative initiative. MethodsIn 2021, general pediatric surgeons representing all five independent university hospitals performing neonatal surgery in Finland initiated national collaboration, the Finnish pediatric surgery hub (FPSH), for sharing of surgical expertise and collective learning. For each case addressed by FPSH, place of care and surgical team were decided individually, and when deemed necessary, operations were performed in cooperation. Operations performed during 2021–2023 and associated early (<30 days) postoperative complications were analyzed according to Clavien-Madadi classification. ResultsOf the total 40 surgeries managed co-operatively by FPSH, 30 (75%) took place in local university hospitals and 10 in Helsinki University Hospital. There were 34 (85%) elective and 6 urgent cases, which were operated within median 1 (range, 1–3) days. Most frequent underlying diagnoses included anorectal malformations, esophageal atresia and Hirschsprung disease. Overall, 12 (30%) had any early postoperative complications, all Clavien-Madadi grade IIIB or lower, and five patients (13%) were reoperated. Rate or grade of complications was not associated with place of care. In addition to regular virtual case meetings, national care protocols and research projects were introduced. ConclusionThese preliminary findings suggest that our national collaborative initiative, FPSH, not only provided practical and safe framework for sharing of surgical expertise but also for collective learning. Level of EvidenceIII.

"The impact of the COVID-19 pandemic on research activities: A survey of the largest Italian academic community".

The aim of the present work is to explore the impact of the COVID-19 pandemic on research activities in a vast multidisciplinary academic community to identify the most critical issues. To this purpose we planned a survey addressed to the entire academic research staff at "Sapienza" University of Rome, which represents the largest Italian academic community. A questionnaire consisting of both open and closed-ended questions was delivered to 4118 individuals in April 2021. A total of 544 responses were collected. All academic roles were sufficiently represented in the study cohort. The median number of critical issues experienced by academic research staff was three. Among these, the three most frequently reported were related to: "Access to libraries / laboratories / research sites" (21.9%), "Limitation to stay abroad / study / research periods" (17.6%), "Progress of experimental work" (14.7%), with variable prevalence according to academic position and gender. Older subjects reported issues with "Projects' financial reporting" and "Expiration of acquired consumable material more frequently". The most common critical aspects reported in relation to the economic burden were: being "Unable to allocate funds" (31.4%), a "Reduction in clinical and scientific activity" (26.3%) and experiencing "Increased expenses (comprising private costs)" (21.2%) with no differences between genders. Researchers in Applied Sciences and Natural Sciences reported a higher frequency of problems in clinical and scientific activities, whereas increased expenses were reported also by researchers operating in the Humanities field. As a possible solution aimed at improving these issues, most subjects, especially those aged >45 years, indicated "Economic aid" (22.6%), "Reduction in bureaucracy" (19.9%) or "Enhancement of the scientific and clinical activities", whereas those aged ≤45 years felt that an increased duration and better access to PhD programs were to be prioritized. Our findings highlight the most critical issues related to research activities during the COVID-19 pandemic in a large academic community. The information achieved may be useful to identify researchers' needs and to design appropriate policies aimed at preparing research institutions for unexpected catastrophic events and limiting the negative impact on academic research activities.

Evaluation of Online Inquiry Competencies of Chilean Elementary School Students: A Dataset

In the age of abundant digital content, children and adolescents face the challenge of developing new information literacy competencies, particularly those pertaining to online inquiry, in order to thrive academically and personally. This article addresses the challenge encountered by Chilean students in developing online inquiry competencies (OICs) essential for completing school assignments, particularly in natural science education. A diagnostic study was conducted with 279 elementary school students (from fourth to eighth grade) from four educational institutions in Chile, representing diverse socioeconomic backgrounds. An instrument aligned with the national curriculum, featuring questions related to natural sciences, was administered through a game named NEURONE-Trivia, which integrates a search engine and a logging component to record students’ search behavior. The primary outcome of this study is a dataset comprising demographic information, self-perception, and information-seeking behaviors data collected during students’ online search sessions for natural science research tasks. This dataset serves as a valuable resource for researchers, educators, and practitioners interested in investigating the interplay between demographic characteristics, self-perception, and information-seeking behaviors among elementary students within the context of OIC development. Furthermore, it enables further examination of students’ search behaviors concerning source evaluation, information retrieval, and information utilization.

Cree-driven community-partnered research on coastal ecosystem change in subarctic Canada: a multiple knowledge approach

Indigenous-driven and community-partnered research projects seeking to develop salient, legitimate, and credible knowledge bases for environmental decision-making require a multiple knowledge systems approach. When involving partners in addition to communities, diverging perspectives and priorities may arise, making the pathways to engaging in principled research while generating actionable knowledge unclear to disciplinarily-trained natural science researchers. Here, we share insights from the Eeyou Coastal Habitat Comprehensive Research Project (CHCRP), an interdisciplinary, Cree-driven community-academic partnership. This project brought together Cree community members, regional organizations, industry (Hydro-Québec), and academics from seven universities across Canada to address the unprecedented loss of seagrass Zostera marina (eelgrass), the concurrent decline in migratory Canada geese and its impact on fall goose harvest activities in Eeyou Istchee. After describing the history and context of the project, we discuss the challenges, complexities, and benefits of the collaborative approach balancing saliency, legitimacy and credibility of the knowledge produced. We suggest the paper may be of use to researchers and partners seeking to engage in principled and actionable research related to environmental change including impacts of past development.

Effectiveness of the Use of Denosumab When Performing Intralesional Surgery in Patients with Gigant Cell Tumors of the Bones of the Knee Joint

In intralesional resection of giant cell tumors, local recurrence rate can be as high as 60–65 %. Some authors attribute recurrences to the presence of an extraosseus component and pathologic fractures, others focus on the effectiveness of bone cavity processing. Currently, denosumab is widely used in treatment of giant cell tumors of the bone. Most frequently, it is used in neoadjuvant regimen at dose 120 mg. This regimen allows to decrease the volume of extraosseous tumor component, form bone margins, and, as a result, make radical resection easier. However, the number of publications on increased risk of local tumor recurrence after neoadjuvant denosumab administration has been growing.Aim. To evaluate the effectiveness of various denosumab administration regimens in intralesional tumor resection in patients with giant cell tumors of the bones forming the knee joint.Materials and methods. The prospective study included outcomes of integrated treatment of 65 patients (34 (52.3 %) women and 31 (47.7 %) men) who underwent surgery at the Division of Bone Oncology of the National Scientific Research Center for Traumatology and Orthopedics named after R.R. Vreden due to giant cell tumors of the knee joint between 2017 and 2023. Mean patient age was 37.5 ± 9.79 years (95 % confidence interval 35.11–39.96; range 19–61 years). Mean follow-up duration was 40.2 ± 13.3 months (95 % confidence interval 36.87–43.46; range 18–68 months). To perform comparative analysis, the patients were divided into 2 groups depending on treatment type. The Group 1 (control) included 32 patients who subcutaneously received 120 mg denosumab prior to surgery (4 injections once a week and then once a month; total number of injections was 6 or more) with subsequent intralesional tumor resection, defect reconstruction with cement in combination with external fixation using a lockable condylar plate. The Group 2 (treatment) included 33 patients who underwent the same surgery as patients of Group 1 and then received subcutaneous injections of 120 mg denosumab (1 injection once a month for 3 months after surgery). Classification and analysis of complications were performed using the International Society of Limb Salvage (ISOLS) 2014 system. Functional outcomes were evaluated using the Musculoskeletal Tumor Society Score (MSTS) at 3, 6, and 12 months.Results. Mean time of recurrence diagnosis in the Group 1 was 9.1 months, in the Group 2 – 14.3 months. Evaluation of the probability of local tumor recurrence depending on treatment type showed statistically significant differences (p = 0.001). Odds of local recurrence in the Group 2 were 10.6 times lower than in the Group 1, and odds of revision surgical intervention in this patient group was 14.5 times higher (with higher probability, this surgery involved oncological endoprosthesis) (p = 0.011). No statistically significant differences in development of distant metastases and probability of death depending on treatment type were found (p = 0.492). According to the obtained data, there are statistically significant differences in functioning of the knee joint per the MSTS scale in the Groups 1 and 2 at 6 and 12 months (p = 0.021 and p = 0.021, respectively). Functional outcomes were better in the Group 2. No significant differences in knee joint functioning per the MSTS scale at 3 months depending on treatment type were found (p = 0.764).Conclusion. Administration of denosumab in the postoperative period allows to improve oncological outcomes due to decreased number of local recurrences (2 (6.1 %) cases), maintenance of excellent local control and good functional results, and to decrease the probability of revision surgical intervention to a minimum. In our study, administration of this drug prior to surgery (intralesional tumor resection) significantly increased the risk of local tumor recurrence (p = 0.001).

The neurophysiological brain-fingerprint of Parkinson’s disease

Research in healthy young adults shows that characteristic patterns of brain activity define individual "brain-fingerprints" that are unique to each person. However, variability in these brain-fingerprints increases in individuals with neurological conditions, challenging the clinical relevance and potential impact of the approach. Our study shows that brain-fingerprints derived from neurophysiological brain activity are associated with pathophysiological and clinical traits of individual patients with Parkinson's disease (PD). We created brain-fingerprints from task-free brain activity recorded through magnetoencephalography in 79 PD patients and compared them with those from two independent samples of age-matched healthy controls (N=424 total). We decomposed brain activity into arrhythmic and rhythmic components, defining distinct brain-fingerprints for each type from recording durations of up to 4min and as short as 30s. The arrhythmic spectral components of cortical activity in patients with Parkinson's disease are more variable over short periods, challenging the definition of a reliable brain-fingerprint. However, by isolating the rhythmic components of cortical activity, we derived brain-fingerprints that distinguished between patients and healthy controls with about 90% accuracy. The most prominent cortical features of the resulting Parkinson's brain-fingerprint are mapped to polyrhythmic activity in unimodal sensorimotor regions. Leveraging these features, we also demonstrate that Parkinson's symptom laterality can be decoded directly from cortical neurophysiological activity. Furthermore, our study reveals that the cortical topography of the Parkinson's brain-fingerprint aligns with that of neurotransmitter systems affected by the disease's pathophysiology. The increased moment-to-moment variability of arrhythmic brain-fingerprints challenges patient differentiation and explains previously published results. We outline patient-specific rhythmic brain signaling features that provide insights into both the neurophysiological signature and symptom laterality of Parkinson's disease. Thus, the proposed definition of a rhythmic brain-fingerprint of Parkinson's disease may contribute to novel, refined approaches to patient stratification. Symmetrically, we discuss how rhythmic brain-fingerprints may contribute to the improved identification and testing of therapeutic neurostimulation targets. Data collection and sharing for this project was provided by the Quebec Parkinson Network (QPN), the Pre-symptomatic Evaluation of Novel or Experimental Treatments for Alzheimer's Disease (PREVENT-AD; release 6.0) program, the Cambridge Centre for Aging Neuroscience (Cam-CAN), and the Open MEG Archives (OMEGA). The QPN is funded by a grant from Fonds de Recherche du Québec - Santé (FRQS). PREVENT-AD was launched in 2011 as a $13.5 million, 7-year public-private partnership using funds provided by McGill University, the FRQS, an unrestricted research grant from Pfizer Canada, the Levesque Foundation, the Douglas Hospital Research Centre and Foundation, the Government of Canada, and the Canada Fund for Innovation. The Brainstorm project is supported by funding to SB from the NIH (R01-EB026299-05). Further funding to SB for this study included a Discovery grant from the Natural Sciences and Engineering Research Council of Canada of Canada (436355-13), and the CIHR Canada research Chair in Neural Dynamics of Brain Systems (CRC-2017-00311).

Specificity of managing the resource potential of scientific and knowledge-intensive organizations in Russia at the present stage of economic development

The article analyzes the problem of creating a national scientific research innovative system at the present stage of economic development, taking into account the Russian specifics of managing the resource potential of scientific and knowledge-intensive organizations. The features of scientific research organizations in the USA, Western Europe and China are examined in detail, global trends in reforming the organization of scientific research institutions are highlighted. The authors conclude about the creation of a national innovative system for organizing scientific research in Russia, promoting the development of an innovative economy.

Corrigendum to: Effects of Long Noncoding RNA AK093407 on the Biological Behavior of Colon Cancer Cells and the Underlying Mechanism

The funding details have been incorporated upon author’s request in the funding section of this article entitled “Effects of Long Noncoding RNA AK093407 on the Biological Behavior of Colon Cancer Cells and the Underlying Mechanism,” 2023, 26(2), 289-300 [1]. &lt;p&gt; Details of the error and a correction are provided here. &lt;p&gt; Original: &lt;p&gt; This work was supported by the important project of the Hebei Province education department (ZD20180004, QN2016017, QN2020208), Hebei Province population and family planning commission technological project (20160009), Chengde Medical University project (202014, 201806). &lt;p&gt; Corrected: &lt;p&gt; This work was supported by The Natural Science Research Youth Fund for Higher Education of Hebei Province (QN2020208), the Hebei Province population and family planning commission technological project (20160009), Chengde Medical University project (202014, 201806), Key Discipline of College and Universities in Hebei Province, China ([2013]04), Technology Innovation Guidance Project-Science and Technology Work Conference (Hebei Provincial Department of Science and Technology). &lt;p&gt; We regret the error and apologize to readers. &lt;p&gt; The original article can be found online at www.eurekaselect.com/article/122378.

Developments in Raman Spectromicroscopy for Strengthening Materials and Natural Science Research: Shaping the Future of Physical Chemistry.

Spectroscopic techniques, especially Raman spectroscopy, cover a large subset in the teaching and research domain of physical chemistry. Raman spectroscopy, and other Raman based techniques, establishes itself as a powerful analytical tool with diverse applications across scientific, industrial, and natural science (including biology and pharmacy) fields and helps in the progress of physical chemistry. Recent advancements and future prospects in Raman spectroscopy, focusing on key areas of innovation and potential directions for research and development, have been highlighted here along with some of the challenges that need to be addressed to prepare Raman based techniques for the future. Significant progress has been made in enhancing the sensitivity, spatial resolution, and time resolution of Raman spectroscopy techniques. Raman spectroscopy has applications in all areas of research but especially in biomedical applications, where Raman spectroscopy holds a great promise for noninvasive or minimally invasive diagnosis, tissue imaging, and drug monitoring. Improvements in instrumentation and laser technologies have enabled researchers to achieve higher sensitivity levels, investigate smaller sample areas with improved spatial resolution, and capture dynamic processes with high temporal resolution. These advancements have paved the way for a deeper understanding of molecular structure, chemical composition, and dynamic behavior in various materials and biological systems. It is high time that we consider whether Raman based techniques are ready to be improved based on the strength of the current era of AI/ML and quantum technology.

Analyzing Science Textbooks In Terms of Inquiry Levels of Activities and The Nature of Scientific Inquiry

This research aims to analyze science textbooks in terms of inquiry levels of activities and the nature of scientific inquiry. The document analysis method is used as a research method. It is found that the nature of the scientific inquiry aspects is not included enough in the textbooks, and some textbook statements can cause students to have misconceptions such as scientific investigations must test a hypothesis, and there is a series of steps that must be followed in all research. It is determined that the activities considered scientific inquiry are mainly at the confirmation level and the structured level in the activities included in the textbooks. It is suggested that more explicit emphasis should be placed on the aspects of the nature of scientific inquiry that are not referred or less referred, the statements confronting these aspects should be removed from the textbooks, and guided and open inquiries should be placed more in the textbooks.

The Valorization of Spanish Minority Grapevine Varieties—The Volatile Profile of Their Wines as a Characterization Feature

Despite the large number of existing varieties of Vitis vinifera L., only few occupy a large niche in today’s highly globalized wine market. The increasing consumer demand for diversified products, as well as the changing climatic conditions, make establishing a process of varietal diversification essential to achieve both challenges. It is for this reason that the study of minority varieties, which have a higher level of adaptation to each area of origin, is of particular interest. With the main objective of achieving an in-depth knowledge of minority varieties in Spain, the national research project ‘Valorization of Minority Grapevine Varieties for their Potential for Wine Diversification and Resilience to Climate Change’ (MINORVIN), has been proposed. Within this extensive project, the present study describes the aroma profiles of 60 single-variety wines, corresponding with 44 different varieties, with 12 of these varieties being studied at the same time in several Spanish regions. Volatile compounds were determined through three consecutive vintages using gas chromatography-mass spectrometry (GC-MS) and gas chromatography–flame ionization detector (GC-FID). Compounds were grouped into major compounds, including alcohols, C6 compounds, esters, acetates, acids, carbonyl compounds, and other type of compounds, and minor compounds, including lactones, terpenes, and C13-norisoprenoids, according to their concentration in the wines being analyzed. Among this last group of compounds, lactones were quantitatively the most abundant, followed by terpenes. This study reflects that minority variety wines show distinctive aromatic profiles, supporting the importance of valuing and promoting the autochthonous minority grapevine varieties for the Spanish winemaking industry.

Dynamic Home Care Routing and Scheduling with Uncertain Number of Visits per Referral

Despite the rapid growth of the home care industry, research on the scheduling and routing of home care visits in the presence of uncertainty is still limited. This paper investigates a dynamic version of this problem in which the number of referrals and their required number of visits are uncertain. We develop a Markov decision process (MDP) model for the single-nurse problem to minimize the expected weighted sum of the rejection, diversion, overtime, and travel time costs. Because optimally solving the MDP is intractable, we employ an approximate linear program (ALP) to obtain a feasible policy. The typical ALP approach can only solve very small-scale instances of the problem. We derive an intuitively explainable closed-form solution for the optimal ALP parameters in a special case of the problem. Inspired by this form, we provide two heuristic reduction techniques for the ALP model in the general problem to solve large-scale instances in an acceptable time. Numerical results show that the ALP policy outperforms a myopic policy that reflects current practice, and is better than a scenario-based policy in most instances considered. Funding: This work was supported by the Natural Sciences and Engineering Research Council of Canada [Grants RGPIN-2018-05225 and RGPIN-2020-210524] and by the Telfer School of Management SMRG Postdoctoral Research Fellowship Support [Grant 2020]. Supplemental Material: The electronic companion is available at https://doi.org/10.1287/trsc.2023.0120 .

Publishing important work that lacks validity or reproducibility – pushing frontiers or corrupting science?

ABSTRACT Scientific research requires objectivity, impartiality and stringency. However, scholarly literature is littered with preliminary and explorative findings that lack reproducibility or validity. Some low-quality papers with perceived high impact have become publicly notable. The collective effort of fellow researchers who follow these false leads down blind alleys and impasses is a waste of time and resources, and this is particularly damaging for early career researchers. Furthermore, the lay public might also be affected by socioeconomic repercussions associated with the findings. It is arguable that the nature of scientific research is such that its frontiers are moved and shaped by cycles of published claims inducing in turn rounds of validation by others. Using recent example cases of room-temperature superconducting materials research, I argue instead that publication of perceptibly important or spectacular claims that lack reproducibility or validity is epistemically and socially irresponsible. This is even more so if authors refuse to share research materials and raw data for verification by others. Such acts do not advance, but would instead corrupt science, and should be prohibited by consensual governing rules on material and data sharing within the research community, with malpractices appropriately sanctioned.

Learning to Optimize Contextually Constrained Problems for Real-Time Decision Generation

The topic of learning to solve optimization problems has received interest from both the operations research and machine learning communities. In this paper, we combine ideas from both fields to address the problem of learning to generate decisions to instances of optimization problems with potentially nonlinear or nonconvex constraints where the feasible set varies with contextual features. We propose a novel framework for training a generative model to produce provably optimal decisions by combining interior point methods and adversarial learning, which we further embed within an iterative data generation algorithm. To this end, we first train a classifier to learn feasibility and then train the generative model to produce optimal decisions to an optimization problem using the classifier as a regularizer. We prove that decisions generated by our model satisfy in-sample and out-of-sample optimality guarantees. Furthermore, the learning models are embedded in an active learning loop in which synthetic instances are iteratively added to the training data; this allows us to progressively generate provably tighter optimal decisions. We investigate case studies in portfolio optimization and personalized treatment design, demonstrating that our approach yields advantages over predict-then-optimize and supervised deep learning techniques, respectively. In particular, our framework is more robust to parameter estimation error compared with the predict-then-optimize paradigm and can better adapt to domain shift as compared with supervised learning models. This paper was accepted by Chung Piaw Teo, optimization. Funding: This work was supported in part by the Natural Sciences and Engineering Research Council of Canada. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2020.03565 .