Perline Aline Delle Demange is a Postdoctoral Fellow at PROMENTA , University of Oslo, Department of Psychology. Her work focuses on the intergenerational transmission of (dis-)advantage using genetically-informed methodologies. Education : PhD in Biological Psychology (2018-2023, Vrije Universiteit Amsterdam); MSc in Cognitive Sciences (2015-2017), Diplôme de l'ENS (2015-2018, both from École Normale Supérieure Paris); BSc in Cell Biology and Psychology of Organisms (2012-2015, Université de Strasbourg). Research Interests : Mechanisms of intergenerational transmission, Mendelian randomization, genetic epidemiology, non-cognitive skills in education, and causal inference in social stratification. Her recent publications investigate genetic and environmental pathways linking family socioeconomic status to educational outcomes, mental health interactions with educational attainment, and methodological comparisons for indirect genetic effects. She maintains active GitHub repositories for open-science implementations and lectures on behavioral genetics.
Yong Chen, PhD, is an Assistant Professor of Biostatistics specializing in advanced statistical methodologies for healthcare data analysis. His research bridges biostatistics, machine learning, and real-world evidence generation with significant contributions to federated learning frameworks and meta-analytic techniques. His core research interests include: Statistical methods for electronic health record data Meta-analysis and network meta-analysis Mathematical statistics with healthcare applications Federated learning for decentralized research networks Causal inference methods for observational data Real-world evidence generation in rare diseases Analysis of Dr. Chen's 2025 publications reveals a dominant focus on privacy-preserving distributed analytics for multi-site healthcare studies, particularly through one-shot lossless algorithms (e.g., COLA-GLM). His work demonstrates methodological innovation in addressing unmeasured confounding through negative control calibration and advancing multitask learning frameworks. Key application areas include SARS-CoV-2 sequelae analysis, neuropsychiatric outcomes, and disparities in obesity treatment access, consistently leveraging electronic health records for real-world evidence generation. Scientific recognition: FACMI (Fellow of the American College of Medical Informatics)
Eric Laber serves as James B. Duke Distinguished Professor of Statistical Science at Duke University, with cross-appointments as Professor of Biostatistics & Bioinformatics in the School of Medicine and Research Professor of Global Health. He co-directs the Duke Computing Initiative and maintains active roles in the Department of Statistical Science. His educational background includes: Ph.D. in Statistics, University of Michigan, Ann Arbor (2011) M.A. in Statistics, University of Michigan, Ann Arbor (2007) Laber's research pioneers statistical reinforcement learning for dynamic decision-making in complex health environments. He specializes in developing methodologies for Sequential Multiple Assignment Randomized Trials (SMARTs) to optimize personalized treatment sequences in precision medicine. His work bridges machine learning with clinical applications, focusing on cancer therapeutics, mental health interventions, and resource-constrained global health settings. Recent innovations address network interference in policy learning and risk-sensitive optimization for digital health platforms. Analysis of his 15 most recent publications reveals strong trends in adaptive clinical trial design, with increasing emphasis on mHealth applications (40% of recent work), PTSD management in cancer survivors (25%), and network-based reinforcement learning (20%). Key methodological threads include Thompson sampling variants (33%), functional data analysis for truncated outcomes (15%), and causal inference under interference (25%). His scientific recognition includes: James B. Duke Distinguished Professorship (premier faculty honor at Duke) Laber secures substantial grant funding for high-impact health research, currently leading 10 active projects totaling over $15M. Major awards include NIH/NCI funding for next-generation SMARTs in cancer therapeutics (2023-2028), NSF support for risk-sensitive statistical learning (2024-2027), and multiple NIH trials for digital mental health interventions. His collaborative approach spans oncology, psychiatry, and global health teams, with special focus on vulnerable populations including cancer survivors and adolescents. As Co-Director of the Duke Computing Initiative, he leads interdisciplinary teams developing scalable computational frameworks for precision medicine. Current efforts integrate reinforcement learning with electronic health records systems to create real-time clinical decision support tools, particularly for resource-limited settings in infectious disease management and chronic pain treatment.
Sahar Abdelnabi is a Principal Investigator at the ELLIS Institute Tübingen and an independent research group leader at the Max-Planck Institute for Intelligent Systems and Tübingen AI Center. She leads the COMPASS (COoperative Machine intelligence for People-Aligned Safe Systems) research group, focusing on developing safe, aligned, and steerable AI agents with emphasis on security, human aspects, and cooperative multi-agent systems. Her research spans the intersection of AI with security, safety, and sociopolitical aspects: Understanding, probing, and evaluating the failure modes of AI models, their biases, emergent risks, and misuse scenarios Designing mitigations, system defenses, white-box control methods, and reasoning enhancements to counter such risks Leveraging AI agents for scientific discovery and advancing society Dr. Abdelnabi's research has identified critical vulnerabilities in AI systems, including being the first to identify, coin, and taxonomize the indirect prompt injection vulnerability in LLM-integrated applications (2023), and proposing watermarking for generative AI (2020). Her work has received significant recognition including a Best Paper Award at ACL2025 and AISec'23 workshop. Her scientific contributions have been widely recognized: Best Paper Award at ACL2025 for "A Theory of Response Sampling in LLMs" Best Paper Award at AISec'23 workshop for "Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection" Oral Paper presentation at ICCV 2021 Spotlight Paper at NeurIPS Datasets and Benchmarks 2024 Dr. Abdelnabi actively mentors students and researchers, hiring across all levels including interns, research visitors, PhD students, and postdocs. She has served on program committees for major AI and security conferences and has presented her work at numerous international venues. Her research has influenced policy and industry practices, with findings adopted by organizations including the German Federal Office for Information Security, NIST, and OWASP. The COMPASS research group provides a dynamic environment for exploring cutting-edge questions in AI safety and security, with opportunities for collaboration across institutions and disciplines.
Jean Lacroix is an Associate Professor at Université Paris-Saclay specializing in political economy and economic history, with additional affiliations as Research Fellow at Université libre de Bruxelles and Research Associate with DIAL. His international engagement includes recent visiting positions at CESifo (May 2025) and prior fellowships at University of Cambridge, Paris School of Economics, and ENS Lyon. His research explores two interconnected agendas: 1) The political economy of war (focusing on WWII France's collaboration networks and long-term societal impacts) and 2) Technology-institution interactions (analyzing how 19th-century innovation shaped political organizations through patent text analysis). This work combines novel historical datasets with rigorous microeconometric methods , particularly difference-in-differences approaches. His recent publication trends reveal increasing sophistication in causal identification within historical contexts, moving beyond descriptive analysis to examine mechanisms like 'indirect connections' in elite survival. The research spans political economy, economic history, and innovation studies with strong methodological rigor. ANR POLECOWWII project: Investigating WWII's long-term effects on trust/voting behavior ANR TECHNOFIRMS project: Analyzing historical patent texts to understand innovation-institution dynamics Professor Lacroix maintains active advising relationships through his ANR-funded projects, providing students with opportunities for archival research, dataset construction, and international collaboration. His lab leverages partnerships with European research centers for comparative historical analysis.
Helene Charlotte Wiese Rytgaard serves as an Associate Professor in the Department of Public Health, Section of Biostatistics at the University of Copenhagen's Faculty of Health and Medical Sciences. Her research focuses on methodological advancements in causal inference, targeted machine learning, event history analysis, and nonparametric estimation. Her primary research interests center on developing statistical machine learning methods for estimating intervention effects in time-varying settings, with specific applications in medical and epidemiological studies. Key areas include survival analysis, competing risks, target population treatment effects, and cumulative incidence estimation. Her work bridges theoretical statistics with practical applications in public health research. Recent publications demonstrate consistent contributions to biostatistical methodology, particularly in targeted maximum likelihood estimation and causal inference frameworks. Her research output shows strong collaboration patterns across epidemiology, psychiatry, and cardiovascular medicine, with significant emphasis on real-world data applications and methodological innovations for complex time-to-event data. Rytgaard maintains active research collaborations across international networks, as evidenced by her extensive co-authorship patterns in high-impact journals. Her work frequently addresses challenges in medical research design and analysis, particularly in observational studies and clinical trials requiring sophisticated statistical approaches.
