Kick off the conference by picking up your badge and connecting with colleagues at the Opening Reception in the Franklin Ballroom. Light refreshments will be served as you meet, mingle, and get ready for an inspiring program ahead. 

Cash Bar available. 

Ease into the day with a Networking Breakfast in the Franklin Ballroom. Enjoy coffee and breakfast while connecting with colleagues before diving into the sessions ahead. 

Join us in the Franklin Ballroom as we officially open the conference. Hear a warm welcome from the organizers and get a preview of what’s ahead in the days to come. 

This talk will introduce knowledge-guided machine learning (KGML), a rapidly growing field in AI for Science where scientific knowledge is deeply integrated in machine learning frameworks to produce scientifically grounded, explainable, and generalizable predictions even on out-of-distribution data. This talk will present a multi-dimensional view to organize prior research in this area and illustrate KGML concepts using a variety of case studies in ecology, biology, and public health including modeling the quality of water in lakes across the US and discovering novel biological traits of organisms linked with evolution from biodiversity images as part of the NSF HDR Imageomics Institute. The talk will conclude with a discussion of how KGML is leading a new paradigm in AI for Science while also advancing the Science of AI driven by the needs of problems in science and engineering.

---

Speaker Bio

Dr. Anuj Karpatne is an Associate Professor of Computer Science at Virginia Tech, where he also serves as a College of Engineering Faculty Fellow and Dean’s Fellow. His research focuses on advancing knowledge-guided machine learning to accelerate discovery in domains including climate science, hydrology, ecology, geophysics, mechanobiology, quantum mechanics, and fluid dynamics. Dr. Karpatne has received multiple awards including the 2025 COE Faculty Fellow Award for Excellence in Research, 2024 NAIRR Pilot Award (with an invited talk at the White House), NSF CAREER Award in 2023, and several early-career faculty awards at Virginia Tech. He is Associate Editor for ACM TKDD, co-author of the second edition of Introduction to Data Mining, and lead editor of the first comprehensive book on Knowledge-guided Machine Learning.

AI is transforming how we do science—not just accelerating research, but expanding the scale of questions we can ask. From molecules to the planet, AI allows us to connect across domains and uncover fundamental insights into how the world works. HDR Institutes bring unique strengths in spanning scales, making it possible to push beyond traditional boundaries and tackle discoveries that weren’t possible before. 

---

Speaker Bios

Tanya Berger-Wolf is a Professor at the Ohio State University, where she is also the Director of the Translational Data Analytics Institute. A pioneer in Artificial Intelligence (AI) for ecology, biodiversity, and conservation, she leads several national and international initiatives and centers and serves as a scientific advisor for many organizations, including the US National Academies Board on Life Sciences, US National Committee for the International Union of Biological Sciences (IUBS), the Global Partnership on AI (GPAI)/OECD, and The Nature Conservancy. Her contributions have earned numerous accolades, including recognition as the AI 100 Global Thought Leaders by H20.ai and most recently the OSU College of Engineering Lumley Interdisciplinary Research Award.

Shih-Chieh Hsu is a Professor of Physics and Adjunct Professor of Electrical and Computer Engineering at the University of Washington. He directs the NSF HDR Institute for Accelerated AI Algorithms for Data-Driven Discovery (A3D3 – https://a3d3.ai). With degrees from National Taiwan University and UC San Diego, Dr. Hsu specializes in experimental particle physics focusing on dark matter searches, neutrino measurements and AI applications in data-intensive research. His work utilizes the Large Hadron Collider and incorporates real-time AI for rapid data analysis and decision-making across multiple scientific disciplines. He has received recognition for his innovative research, mentorship, and contributions to real-time AI applications in scientific discovery. 

Vandana Janeja is Associate Dean for Research and Faculty Development in the College of Engineering and Information Technology, Professor of Information Systems department at the University of Maryland Baltimore County (UMBC).  She is the director of iHARP, an NSF HDR Institute for Harnessing Data and Model Revolution in the Polar Regions.  Her research is in the area of data science and AI with a focus on spatio-temporal analytics, fine grained computer vision algorithms, physics driven machine learning, modeling heterogeneity across multiple domain datasets. Her work looks at societal impacts such as climate change, ethics in data science, misinformation detection and security through the lens of her research in data science and AI.    

Anand Padmanabhan is a Research Associate Professor in the Department of Geography and Geographic Information Science at the University of Illinois at Urbana-Champaign (UIUC) and is the academic advisor for the Online MS program in CyberGIS and Geospatial Data Science. He holds a Ph.D. degree in computer science from the University of Iowa and has research interests in advanced cyberinfrastructure, geographic information science and systems (GIS), cyberGIS, and geospatial data science. Specifically, his research focuses on democratizing access to advanced cyberinfrastructure for enabling geospatial discovery and innovation by developing and operating cyberGIS capabilities and services. He is the Managing Director of the Institute for Geospatial Understanding through an Integrative Discovery Environment (I-GUIDE) funded by the National Science Foundation and serves as a co-lead for its Cyberinfrastructure team.​ He has served as an investigator for several projects funded by the U.S. National Science Foundation.

A buffet lunch will be served in the Franklin Ballroom for all conference participants. Early career faculty are invited to bring their lunch to the Ohio Ballroom for a mentorship session with senior faculty hosted by ID4. 

This dynamic session features a series of rapid-fire presentations where researchers have just five minutes to share their work. Designed to spark curiosity and highlight a wide range of ideas across the HDR Ecosystem, these talks provide a fast-paced glimpse into emerging projects, novel approaches, and future directions. Join us to discover fresh perspectives and connect with colleagues pushing the boundaries of research. 

---

Lightning Talks

Leveraging AI to extract high resolution morphometric data and discover new bat species – Bryan Carstens & Sydney Decker, Evolution, Ecology, Organismal Biology, The Ohio State University

Convolutional neural networks are implemented to extract high resolution data from bat skulls. The CNNs were used to automate a pseudo-landmarking procedure on a three dimensional reconstruction of bat dentary and cranial bones which were obtained via micro CT scanning. These data were then used to quantify subtle shape differentces among yellow bat specimens and confirm preliminary species boundaries originally suggested by genomic data. Our results demonstrate that imageomic tools can be a key addition integrate morphometric data into systematic biology and taxonomy.

Towards Knowledge-Centric Generative Modeling – Cheng Zhang, Computer Science and Engineering, Texas A&M University

Recent advances in generative models have transformed image and text synthesis, yet their use in scientific domains often overlooks structured knowledge that is central to reliable discovery. This talk introduces knowledge-centric generative modeling that incorporates external sources such as taxonomic hierarchies and domain rules to guide generation. I will share examples from our work on text-to-image models for animal species and knowledge-informed image generation pipelines, showing how explicit priors can improve reliability, interpretability, and inclusiveness.

Real-time AI for Neuroscience – Amy Orsborn, Electrical & Computer Engineering, University of Washington

The goal of systems neuroscience is to understand how brain networks perform computations to control our behavior. Neuroscience experiments dissect these relationships using inferences between behavioral measurements and neural recordings, and interventions that reveal causal links between neural activity and function. Technological advances have rapidly expanded our ability to measure neural activity and behavior, ushering in a new regime where neuroscience can leverage machine learning. Applying ML in real-time also enables closed-loop interventions, such as brain-computer interfaces, which are increasingly valuable tools to causally interrogate brain-behavior relationships and provide paths to therapies. My talk will briefly summarize the A3D3 HDR Institute’s work to develop low-latency, high throughput machine learning tools for neuroscience data, with a focus on recent work to leverage transformers and “foundation model” approaches to create models capable of zero or few-shot generalization to new data, which are critical for real-time experiments. 

Evaluation & Knowledge Transfer of Convergence Science – Diana Sinton, NA, University Consortium for Geographic Information Science

The same characteristics of convergence science that make its approach compelling, i.e. deliberate integration of diverse fields to address complex problems and produce more holistic results, generate both challenges and opportunities in the areas of evaluation and knowledge transfer. Examples will come from I-GUIDE’s multiple research activity areas. 

Explainable AI and Evidential Deep Learning for Anomaly Detection – Mark Neubauer, Physics, University of Illinois at Urbana-Champaign

In this talk, I will discuss explainable AI methods and evidential deep learning (EDL) applied to the identification of jets in high-energy proton-proton collisions at the Large Hadron Collider. I will also touch on future directions to leverage EDL-based uncertainty quantification for improved anomaly detection. 

It's ARK time: the data deluge is here – Matthew Graham, Cahill Center for Astronomy and Astrophysics, California Institute of Technology

With the Rubin Observatory now in its science validation phase, the era of massive real-time data streams in astronomy is here. Fast ML is essential to ensuring optimal scientific discovery and we'll give a brief review of how we are building an Accelerated Revelation of the Kosmos. 

The Evolution of Wing Coloration in Lepidoptera – Moritz Luerig, Florida Museum of Natural History, University of Florida

Butterfly and moth wings show an astonishing diversity of color patterns, yet these are often controlled by just a few genes. My research explores how such a simple developmental system can give rise to such rich phenotypic and taxonomic diversity. Using AI-based image analysis, I’m building a large-scale database of wing coloration and shape traits from museum collections and citizen science platforms. Funded by the European Research Council, this project will help trace the evolutionary history of wing patterns, link them to species diversification, and test how environmental factors shape global color variation. 

Accelerating Gravitational-Wave Astronomy with Machine Learning – Bhavya Gupta, Department of Physics and LIGO Laboratory, Massachusetts Institute of Technology

With the growing volume and complexity of gravitational-wave (GW) data, machine learning and artificial intelligence are shaping the analysis era of GW astronomy. Several dedicated frameworks, such as ML4GW, including methods like Aframe and AMPLFI, have been developed to enable end-to-end searches, from signal detection for compact binary coalescences to rapid parameter estimation. Achieving low-latency performance is especially critical for multi-messenger astronomy, where rapid sky localization is key to enabling timely electromagnetic follow-up of GW events. I will highlight progress in building such low-latency ML frameworks, discuss results from public searches on LIGO data, and outline future goals to strengthen existing pipelines by advancing real-time detection and enhancing the prospects for multi-messenger astronomy. 

Morphological Barcoding: A novel method of extracting diagnostic trait data from images of organisms – Hank Bart, Biodiversity Research Institute, Tulane University

Morphological Barcoding, a novel method of extracting diagnostic trait data from images of organisms, that grew out of research for the Imageomics Institute, has the potential to greatly accelerate the pace of biodiversity discovery and eliminate the taxonomic Impediment 

DimBridge: Interactive Explanation of Visual Patterns in Dimensionality Reductions with Predicate Logic – Mingwei Li, Computer Science, Tufts University

Dimensionality reduction techniques are widely used to make sense of high-dimensional data by turning it into easier-to-view two-dimensional projections. However, there is often limited support for connecting these projections back to the original data, which can make it hard for users to gain clear insights. We introduce DimBridge, a visual analytics tool that lets users explore patterns directly in the projection and see the related patterns in the original data. DimBridge supports a variety of tasks, such as comparing clusters or breaking down more complex structures. Using first-order predicate logic, it finds the subspaces most relevant to a user’s query and provides an interface for exploring and interacting with them. We show how DimBridge can make it easier for users to understand and interpret visual patterns in projections. 

Beginning to understand the requirements for AI-ready ecology and biodiversity data infrastructure for open science – Eric Sokol, National Ecological Observatory Network (NEON), Battelle

Over the past several decades, there has been a rapid increase in open and FAIR ecological and biodiversity data. This has partly been driven by intentional, national-scale efforts to fund and coordinate the collection and publication of open data, including programs such as the US Long Term Ecological Research Network (US LTER), the National Ecological Observatory Network (NEON), NatureServe, and many others. During this time, there has also been a proliferation in more localized efforts to collect monitoring data (e.g., California’s Surface Water Ambient Monitoring Program (SWAMP)), growth in datasets produced by citizen science (e.g., iNaturalist and eBird), and efforts to digitize museum collections (e.g., those coordinated by iDigBio). This continually expanding wealth of data in combination with recent advances in technology (AI and ML tools) provide an unprecedented opportunity in human history to understand species and ecosystems. However, barriers remain to unlocking this potential because of the heterogeneity of the data landscape, siloes that exist within and among disciplines, fragmented infrastructure, limited interoperability, and other challenges yet to be identified. By bringing together a diverse, interdisciplinary team from the life sciences and computer sciences, we have begun to inventory data sources, AI-critical use cases, and barriers to implementation with a goal of designing a prototype data infrastructure. Ultimately, our goal is to identify requirements for an AI-ready infrastructure that enable development of AI-enabled approaches to advance ecological science and decision-making. 

Automated Knowledge Extraction from Scientific PDFs using LLMs and Graph-based Techniques – Ajith Kumar Dugyala, Data Science, University of North Texas

The variations in how articles are organized, the multifaceted nature of the information, and the specialist language of these articles make it hard to examine extensive PDF collections of research articles by non-experts of that domain area. Since there is no consistent structure, automatic methods cannot collect, review or display key concepts clearly, particularly in science, where connecting ideas is essential. In light of above challenges, we propose a framework to expedite interdisciplinary research, automate the combination of ideas and gain helpful insights from growing scholarly works. 

Overview of the National Artificial Intelligence Research Resource (NAIRR) Pilot - Shelley Knuth, University of Colorado Boulder

The National Artificial Intelligence Research Resource (NAIRR) is a federally supported initiative designed to democratize access to advanced AI infrastructure, datasets, and expertise across the U.S. research and education community. Its mission is to lower barriers to entry for AI research, ensuring that everyone can participate fully in the AI revolution. By federating high-performance computing, cloud resources, curated datasets, and training programs, NAIRR aims to accelerate discovery, expand the talent pipeline, and strengthen U.S. leadership in AI. This talk will provide a brief overview of NAIRR’s goals, lessons learned, and opportunities for researchers, educators, and institutions to engage.

Real-Time Adaptive Control Architecture for Physics and Neuroscience Applications – Jai Yu, Psychology, University of Chicago

Adjusting the operation of a device very fast and in real-time based on live information from sensor signals has important applications across research disciplines. Machine learning-based algorithms have the potential to use complex inputs to aid real-time signal analysis. We are working on control architectures that can perform fast signal processing with added self-monitoring to adapt to changes in the signal real-time. Further, we aim to deploy these control algorithms on edge devices with computational constraints. We aim to use these control systems for physics and neuroscience applications.

This discussion explores how scientific frontiers are shaped and resourced, with perspectives from both the National Academies of Sciences, Medicine, and Engineering (NASEM) and the National Science Foundation (NSF). NASEM will share how it identifies emerging questions at the frontiers of AI-enabled science—highlighting recent efforts and what lies ahead. NSF will discuss how these opportunities are translated into research investments, from funding priorities to the challenges of resourcing AI-intensive science. Together, we will examine how ideas move from vision to implementation at the cutting edge of discovery. 

---

Speaker bios

Kavita Berger is the Board Director of the Board on Life Sciences and co-Director of the Board on Animal Health Sciences, Conservation, and Research of the National Academies of Sciences, Engineeering, and Medicine. She is a life scientist with extensive experience in the addressing a diversity of technical, policy, national security, and societal issues associated with the life sciences and biotechnology. Dr. Berger leads and oversees the Boards’ work across a variety of life science areas, including basic, applied, and emerging life sciences research; transdisciplinary biotechnology research and convergence; bioeconomy-related research and development; biosecurity and biodefense; ecology and biodiversity; integrated human, animal, plant, and ecological health; and animal health. A list of Dr. Berger’s publications is accessible through her MyNCBI bibliography. Dr. Berger has a Ph.D. in genetics and molecular biology from Emory University and conducted pre-clinical research on HIV vaccines.

Chaitan Baru has been Senior Advisor since 2022, and is currently Acting Section Head for Emerging Technologies, in the Technology, Innovation, and Partnerships (TIP) Directorate at NSF. He first joined NSF (2014-2018) as Senior Advisor for Data Science in the CISE Directorate where he co-chaired the NSF Harnessing the Data Revolution (HDR) Big Idea, led the BIGDATA program, and initiated the HDR Data Science Corps program. He returned to NSF from 2019-2022 as Senior Advisor for the NSF Convergence Accelerator and was a member of the team that launched the first several “tracks” of the program, including the Open Knowledge Network, a component of the original HDR vision. Prior to joining NSF, Chaitan had a 25-year career at the San Diego Supercomputer Center, University of California San Diego, where he held a variety of leadership positions. His prior experience also includes positions at IBM and in the EECS Department at the University of Michigan.

Artificial intelligence is increasingly positioned as a partner in scientific discovery. Yet the promise of AI is entangled with challenges of trust, interpretability, and overreliance. This talk examines how people use AI in data-driven decision-making, with particular attention to balancing the benefits of algorithmic suggestions against the need for human critical thinking. Drawing on empirical work at the intersection of visualization and human–AI interaction, I will discuss strategies for designing explanations and transparency mechanisms that both encourage productive use of AI and help users calibrate their reliance. While large language models increasingly dominate discussions of AI, I will situate them within a broader trajectory and highlight emerging experiments with LLMs in visualization contexts as a means to reflect on future directions. The talk concludes by considering how the scientific community can move toward AI systems that not only accelerate discovery but also cultivate more rigorous, transparent, and trustworthy reasoning with data.

---

Keynote Speaker Bio:

Alvitta Ottley is an Associate Professor of Computer Science & Engineering at Washington University in St. Louis, with courtesy appointments in Psychological & Brain Sciences and the Sam Fox School of Design. Her research combines computer science, psychology, and design to improve how people make decisions with data. She develops human-in-the-loop visual analytics systems that align with how people think, focusing on transparency, trust, and effective human–AI collaboration. Dr. Ottley received the NSF CRII Award (2018) for visualization in medical decision-making, the NSF CAREER Award for context-aware visual analytics, and the 2022 EuroVis Early Career Award. She is a Co-PI of the NSF Institute for Data-Driven Dynamical Design (ID4), where her work advances human–AI teaming in scientific discovery, and the lead PI of Carbon Utilization Redesign for Biomanufacturing (CURB), an NSF Engineering Research Center project on carbon utilization for biomanufacturing. Her work appears in top visualization venues, including ACM CHI and IEEE VIS, earning multiple best paper awards. 

Ease into the day with a Networking Breakfast in the Franklin Ballroom. Enjoy coffee and breakfast while connecting with colleagues before diving into the sessions ahead. 

Start Day 2 in the Franklin Ballroom with a warm welcome and important announcements. Get oriented for the day ahead and hear updates to make the most of your conference experience. 

Join us for the official launch of the second HDR ML Challenge. This year’s FAIR challenge program introduces three new scientific benchmarks—Neural Forecasting, Climate Prediction using Ecological Data, and Coastal Flooding Prediction over time—pushing the boundaries of modeling out-of-distribution data. Machine learning problems are often driven by the quality of the available training datasets. Models are very effective at interpolating across their training datasets to find patterns and trends. In this challenge, we ask models to extend beyond their training by performing out of domain extrapolation to practical critical scientific process that have not yet been well studied. Following an introduction to this year’s challenge, we will dedicate time to a community-guided Q&A/Networking session. 

Josephine Namayanja is a Research Associate Professor at the Institute for Harnessing Data and Model Revolution in Polar Regions (iHARP), at the University of Maryland, Baltimore County. Before that, she served as an Assistant Professor of Management Science and Information Systems in the College of Management at the University of Massachusetts Boston. Josephine received a Ph.D in Information Systems at the University of Maryland, Baltimore County in May 2015 where she also received an M.S. in Information Systems in May 2010. She received a B.S. in Information Technology from Makerere University Kampala in Uganda in May 2007.

Elizabeth Campolongo is the Senior Data Scientist for both the Imageomics Institute and AI & Biodiversity Change (ABC) Global Center, based at The Ohio State University. She specializes in large-scale data analysis and preparation, with particular expertise in identifying, evaluating, and preparing biodiversity data for AI/ML model training. Central to her work is ensuring FAIR and reproducible research products. To this end, she has led the creation of web-based guides to FAIR and collaborative science for interdisciplinary teams and co-created FAIR data access and validation software to facilitate adherence to these standards. She is passionate about open-science and has been a co-lead on various workshops and challenges designed to engage the broader community in ML for science pursuits. She earned her Ph.D. in Mathematics from The Ohio State University in 2022. She earned her Ph.D. in Mathematics from The Ohio State University in 2022 under the supervision of Dr. Krystal Taylor.

Amy Orsborn is Clare Boothe Luce Assistant Professor in Electrical & Computer Engineering and Bioengineering at University of Washington. She works at the intersection of engineering and neuroscience to develop therapeutic neural interfaces. She completed her Ph.D. at the UC Berkeley/UCSF Joint Graduate Program in Bioengineering developing co-adaptive strategies for brain-machine interfaces where machine-learning and neural adaptation collaborate to improve system performance. In her postdoctoral training at NYU’s Center for Neural Science, she developed novel neural implants for multi-scale, multi-modal interrogation and monitoring of neural circuits in non-human primates. These implants enable new ways to study neural mechanisms of learning in large-scale networks. Her work has been supported by NSF Graduate Research Fellowship, a pre-doctoral award from the American Heart Association, and a L’Oreal USA for Women in Science postdoctoral award. 

Jump into quick, meaningful connections during our Speed Networking session. Dive deeper into the HDR Challenge topics, exchange ideas with fellow participants, and set the stage for future collaborations. 

Scientific research increasingly relies on combining data from diverse sources—different formats, scales, and domains. This session explores strategies for managing, integrating, and analyzing heterogeneous data, from technical solutions and standards to workflows and tools that enable discovery across complex datasets. Panelists will share challenges, best practices, and emerging approaches to making sense of diverse data in ways that advance research and collaboration. 

---

Moderator Bio

Josephine Namayanja is a Research Associate Professor at the Institute for Harnessing Data and Model Revolution in Polar Regions (iHARP), at the University of Maryland, Baltimore County. Before that, she served as an Assistant Professor of Management Science and Information Systems in the College of Management at the University of Massachusetts Boston. Josephine received a Ph.D in Information Systems at the University of Maryland, Baltimore County in May 2015 where she also received an M.S. in Information Systems in May 2010. She received a B.S. in Information Technology from Makerere University Kampala in Uganda in May 2007.

Speaker Bios

Vandana Janeja is Associate Dean for Research and Faculty Development in the College of Engineering and Information Technology, Professor of Information Systems department at the University of Maryland Baltimore County (UMBC).  She is the director of iHARP, an NSF HDR Institute for Harnessing Data and Model Revolution in the Polar Regions.  Her research is in the area of data science and AI with a focus on spatio-temporal analytics, fine grained computer vision algorithms, physics driven machine learning, modeling heterogeneity across multiple domain datasets. Her work looks at societal impacts such as climate change, ethics in data science, misinformation detection and security through the lens of her research in data science and AI.    

Carol Song is the chief scientist at Purdue University’s Rosen Center for Advanced Computing. With a background in high-performance computing, data infrastructure, and software engineering, she has spent the past 15 years connecting domain research with advanced computing through cyberinfrastructure innovations. She leads the NSF-funded Anvil HPC system and directs the Purdue Center for Research Software Engineering, overseeing multiple large projects that develop user-focused software and platforms for computational and data-driven science. As Co-PI of the NSF HDR I-GUIDE Institute, Carol co-leads the cyberinfrastructure team that builds its geospatial data science platform. Prior to Purdue, she led R&D in medical imaging, networks, and mobile computing in industry. Carol holds a Ph.D. in computer science from the University of Illinois at Urbana-Champaign.

Shaowen Wang is a Professor in the Department of Geography and Geographic Information Science and the Siebel School of Computing and Data Science at the University of Illinois Urbana-Champaign (UIUC). He serves as Associate Dean for Life and Physical Sciences in the College of Liberal Arts and Sciences and as a Senior Faculty Fellow in the Office of the Vice Chancellor for Research and Innovation at UIUC. He is the founding director of UIUC’s CyberGIS Center for Advanced Digital and Spatial Studies and leads the National Science Foundation–funded Institute for Geospatial Understanding through an Integrative Discovery Environment (I-GUIDE). His research focuses on advancing cyberGIS, geospatial data science, and spatial artificial intelligence (AI) to develop scalable solutions for complex geospatial problems and sustainability challenges. He is a Fellow of the American Association for the Advancement of Science, the American Association of Geographers, and the University Consortium for Geographic Information Science.

This session opens with a talk from Bruce Weinberg on the future of AI in research. and the skills tomorrow’s scientists will need to stay competitive. He’ll explore how domain expertise remains essential, what new competencies are emerging, and the big question of whether AI will replace certain roles—or transform them. Following his talk, a panel discussion will dive deeper into these themes, offering diverse perspectives on how students and postdocs can position themselves to thrive in the evolving research workforce. 

---

Moderator Bio
Dr. Tanya Berger-Wolf is a Professor at the Ohio State University, where she is also the Director of the Translational Data Analytics Institute. A pioneer in Artificial Intelligence (AI) for ecology, biodiversity, and conservation, she leads several national and international initiatives and centers and serves as a scientific advisor for many organizations, including the US National Academies Board on Life Sciences, US National Committee for the International Union of Biological Sciences (IUBS), the Global Partnership on AI (GPAI)/OECD, and The Nature Conservancy. Her contributions have earned numerous accolades, including recognition as the AI 100 Global Thought Leaders by H20.ai and most recently the OSU College of Engineering Lumley Interdisciplinary Research Award.

Speaker Bios

Kavita Berger is the Board Director of the Board on Life Sciences and co-Director of the Board on Animal Health Sciences, Conservation, and Research of the National Academies of Sciences, Engineeering, and Medicine. She is a life scientist with extensive experience in the addressing a diversity of technical, policy, national security, and societal issues associated with the life sciences and biotechnology. Dr. Berger leads and oversees the Boards’ work across a variety of life science areas, including basic, applied, and emerging life sciences research; transdisciplinary biotechnology research and convergence; bioeconomy-related research and development; biosecurity and biodefense; ecology and biodiversity; integrated human, animal, plant, and ecological health; and animal health. A list of Dr. Berger’s publications is accessible through her MyNCBI bibliography. Dr. Berger has a Ph.D. in genetics and molecular biology from Emory University and conducted pre-clinical research on HIV vaccines.

Alex Davis is Chief Data Scientist at The Ohio State University’s Translational Data Analytics Institute (TDAI), where he leads Data Science Services, a team that partners with researchers across disciplines to apply advanced data analytics and machine learning methods. With a PhD in Physics, Alex specializes in large-scale data integration, natural language processing, and the application of artificial intelligence to research problems in public health, medicine, ecology, and beyond. He has experience building data infrastructure to support interdisciplinary collaborations and is committed to helping researchers adopt data science practices that enhance rigor, reproducibility, and impact. His current projects include extracting morphometrics from roaches to exploring applications of large language models in research contexts, and fostering AI readiness across the university. 

Jenny Grabmeier, MA, is Director of Research Strategy and Team Science facilitator for Ohio State University’s Translational Data Analytics Institute, OSU’s largest interdisciplinary research institute which focuses on AI and data science approaches to complex challenges across science domains. She serves as team science facilitator for the NSF HDR Imageomics Instituteand NSF- and NSERC-funded Artificial Intelligence and Biodiversity Change (ABC) Global Center, is key personnel for the NSF FAIROS AI-Ready Ecology and Biodiversity Data Infrastructure for Science and Action, and manages the AI and Health research grant program for the NSF-NCATS-funded Clinical and Translational Science Institute at OSU. Jenny specializes in interdisciplinary scientific teaming, knowledge integration and translational processes. She is a published contributor to the science of team science, and a co-instructor of a graduate-level course on Interdisciplinary Team Science.

Bruce A. Weinberg is Eric Byron Fix-Monda Endowed Professor of Economics and Public Affairs at Ohio State University. His research, which has been published in journals including the American Economic Review, Journal of Political Economy, Nature, PNAS, and Science, spans three areas: (1) The economics of creativity and innovation (2) The determinants of youth outcomes and behavior and (3) Technological change, industrial shifts, and the wage structure. He has held visiting positions at the Hoover Institution at Stanford University, the National Bureau of Economic Research (NBER), and Princeton University and is a Research Fellow at the IZA Institute for Labor, Bonn and a Research Associate at the NBER. He has advised over 100 undergraduate and graduate students and postdocs and was awarded Ohio State’s Postdoctoral Mentor of the Year in 2022. He has applied his expertise on science and innovation into practice, having advised the NIH Directorate and the American Association for the Advancement of Science. He has received Ohio State’s Joan Huber Research and Distinguished Scholar Awards and is an Elected Fellow of the AAAS.

A participant-led forum where you set the agenda. Bring your ideas, join conversations that matter to you, and connect with others in a dynamic, collaborative space. 

A buffet lunch will be served in the Franklin Ballroom for all conference participants. Students and postdocs are invited to bring their lunch to the Ohio Ballroom for an informal opportunity to connect and engage with one another, hosted by the Imageomics Institute.

This panel explores innovative approaches to preparing students, early-career researchers, and educators for a world where AI is central to teaching, learning, and research. Panelists will discuss how generative AI is currently being used in educational and research contexts, and strategies for building data literacy and AI fluency at all levels. The session will also highlight novel training programs developed within the HDR ecosystem and share best practices for fostering collaboration, inclusivity, and responsible use of AI across academic and industry settings. 

---

Moderator Bio

Dr. Leanna House is an Associate Professor of Statistics at Virginia Tech (VT), Blacksburg, Virginia.  Prior to her appointment at VT, she earned her M.A.T. in Curriculum Development from Cornell University, Ithaca, New York; earned her Ph.D. and M.S. in Statistics from Duke University, Durham, North Carolina; and completed a postdoc at The University of Durham, U.K.    In addition to serving as a professor at Virginia Tech, Dr. House is a Deputy Division Leader for Computational Methods and Data Analytics (CMDA) and Director of Education and Community for the recently funded Imageomics Institute (imageomics.org).   Dr. House’s research interests include Bayesian statistical modeling with an emphasis in visualization and uncertainty quantification, as well as, analytic methods that foster human-computer interaction and education.  

Speaker Bios

Shelley Knuth is the Assistant Vice Chancellor for Research Computing at the University of Colorado Boulder. She oversees advanced computing and data services that support researchers nationwide, including supercomputing, large-scale data storage, secure enclaves, and high-speed networking. She also serves as Executive Director of the Center for Research Data and Digital Scholarship (CRDDS) and chairs the Rocky Mountain Advanced Computing Consortium (RMACC), fostering collaboration across the region. Shelley is the lead principal investigator for the NSF-funded ACCESS Support project and contributes to several other NSF initiatives. Additionally, she helps guide national strategy as co-lead of the User Experience Working Group for the National Artificial Intelligence Research Resource (NAIRR) pilot. She earned her PhD in Atmospheric and Oceanic Sciences from CU Boulder in 2014.

Hilmar Lapp works at the intersection of biology, computer science, and software engineering. He currently serves as co-PI of the HDR Imageomics Institute and is part of its leadership team in the role of Director of Data and Informatics Infrastructure. Throughout his career, he has initiated and led initiatives and organizations promoting FAIR, reproducible, and open science coupled with workforce training and community capacity building. He is a co-founder of Data Carpentry, which teaches fundamental data skills to researchers; initiated code fests and data paloozas, which empower communities with know-how and collaborative networks; and contributed to community standards bodies to build shared practices. His research program, which includes leading roles in Phenoscape from its inception, revolves around using symbolic and neural AI to extract fully computable organismal traits and clade definitions from unstructured biological data such as natural language text descriptions and biodiversity images.

Stephen Moysey is Director of the Water Resources Center and Professor in the Department of Earth, Environment, and Planning at East Carolina University. Trained as a hydrogeophysicist, he pioneered the use of neural networks for pattern recognition in ground-penetrating radar and has advanced innovative approaches to model–data integration in hydrology. His education research spans the use of mobile devices, game-based learning, virtual reality, and experiential learning, and he has led multiple programmatic grants, including an NSF NRT focused on integrating data science research with community applications. Committed to broadening participation in science, Dr. Moysey is currently investigating how large language models (LLMs) can lower barriers between non-scientists and scientific knowledge. He is also the founder of PersonAIlized, an AI-education company combining LLMs with learning theory to create next-generation adaptive learning and assessment platforms.

Diana Sinton is a Senior Research Fellow with the University Consortium for Geographic Information Science (UCGIS), a non-profit organization based in the United States that supports a community of practice around GIScience in higher education. Prior to her current role, she served UCGIS as its Executive Director for eight years, overseeing all organizational activities including the digital expansion of the UCGIS Geographic Information Science & Technology Body of Knowledge (GIS&T BoK).  For UCGIS Diana contributes to externally-funded projects such as I-GUIDE, for which she serves as a project manager and team leader for evaluation and knowledge transfer.  She also teaches courses in spatial analysis and GIS as an adjunct associate professor at Cornell University (New York, USA). 

Brooke A. Whitworth is a Professor of Science Education at the University of South Carolina, where her research focuses on the roles of district science coordinators, professional development, and science teacher leadership. Before joining USC, she was an Associate Professor and Assistant Chair of Teaching & Learning at Clemson University. She also held Assistant Professor positions at the University of Mississippi and Northern Arizona University. Whitworth’s K-12 teaching career spanned nine years, including roles as a chemistry, physics, and math teacher at the University North Carolina School of the Arts and Punahou School. An active leader in the field, she serves as the Treasurer-Secretary for the National Association for Research in Science Teaching (NARST) and is the editor for the National Science Teaching Association's (NSTA) publication, The Science Teacher. Her contributions have been recognized with the 2023 Association for Science Teacher Education Outstanding Science Educator Award (first 10 years) and appointment as a 2023 NSTA Fellow.

A participant-led forum where you set the agenda. Bring your ideas, join conversations that matter to you, and connect with others in a dynamic, collaborative space. 

How do deep learning vision models arrive at their predictions? This fundamental question has motivated a wide range of explainable and interpretable AI methods, such as saliency methods that highlight where a model looks.

In this talk, Dr. Chao will provide a brief introduction to explainable and interpretable AI, including popular approaches such as Grad-CAM and concept bottleneck models. He will then focus on explainable and interpretable AI for Imageomics (https://imageomics.osu.edu/), especially on highlighting the visual traits that distinguish different species. Specifically, Dr. Chao will present his research team's recent advances in finer-grained CAM (Finer-CAM), interpretable transformers, sparse autoencoder (SAE) for scientific discovery, and other ongoing projects. He will conclude with a discussion on the future research opportunities in this exciting frontier.

---

Speaker Bio

Wei-Lun (Harry) Chao is an Associate Professor in Computer Science and Engineering and a College of Engineering Innovation Scholar at The Ohio State University (OSU). His research focuses on machine learning and computer vision, with applications spanning visual recognition, autonomous driving, biology, and healthcare. He aims to develop fundamental understandings and robust, widely applicable algorithms to tackle real-world challenges. He is particularly interested in learning from imperfect data, including limited, noisy, heterogeneous, distribution-shifting, and inaccessible data. His contributions have been recognized by several awards and honors, including the OSU Early Career Distinguished Scholar Award (2025) and CVPR Best Student Paper Award (2024). Before joining OSU in 2019, he was a Postdoctoral Associate at Cornell University (2018–2019), working with Kilian Weinberger and Mark Campbell. He earned his Ph.D. in Computer Science from the University of Southern California (2013–2018) under the supervision of Fei Sha. 

Generative AI is transforming how researchers access, analyze, and communicate knowledge. This interactive workshop provides a high-level introduction to the rapidly evolving landscape of generative AI, with a focus on its applications in research. Participants will gain an overview of large language model (LLM) technology, explore current platforms and tools, and consider critical issues of security and ethics. The session will also include guided, hands-on activities, covering the fundamentals of prompt engineering in a research context. The workshop will demonstrate working with commercial LLMs and Google’s Notebook LM to practice effective strategies for generating insights, refining outputs, and integrating these tools into their own research workflows. This workshop is designed for researchers across disciplines who are interested in understanding both the potential and the limitations of generative AI.

---

Facilitator Bio

Alex Davis is Chief Data Scientist at The Ohio State University’s Translational Data Analytics Institute (TDAI), where he leads Data Science Services, a team that partners with researchers across disciplines to apply advanced data analytics and machine learning methods. With a PhD in Physics, Alex specializes in large-scale data integration, natural language processing, and the application of artificial intelligence to research problems in public health, medicine, ecology, and beyond. He has experience building data infrastructure to support interdisciplinary collaborations and is committed to helping researchers adopt data science practices that enhance rigor, reproducibility, and impact. His current projects include extracting morphometrics from roaches to exploring applications of large language models in research contexts, and fostering AI readiness across the university. 

Put your work in the spotlight during the Poster Session. Highlight your insights, exchange ideas, and connect with colleagues across the community. 

👉 Register for the Poster Session 

Cash Bar available. 

Gather with fellow attendees for a relaxed evening featuring an Italian buffet. Enjoy great food, lively conversation, and the chance to connect outside the day’s sessions. 

Vegan/Vegetarian options available. Cash Bar available. 

Kick off Day 3 in the Franklin Ballroom with a warm welcome and key announcements. Get set for the final day of the conference and hear the latest updates to help you make the most of your experience. 

Data is at the heart of HDR research, but ensuring its long-term persistence, accessibility, and usability remains a critical challenge. This panel will address strategies for sustaining data resources over time—from infrastructure and funding models to community practices that promote stewardship and reuse. Panelists will share perspectives on how to balance innovation with sustainability, and how the HDR community can work together to safeguard data as a lasting resource for discovery. 

---

Moderator Bio

Dr. Paula Mabee has served as the Chief Scientist and Observatory Director for the National Ecological Observatory Network (NEON) since early 2020. NEON is a 30-year National Science Foundation-funded project that collects standardized data at 81 field sites across the U.S. to advance understanding and forecasting of the complex interactions between ecosystems and the environment by providing high-quality, long-term ecological data and infrastructure to users. Paula is a Distinguished Emeritus Professor at the University of South Dakota, where she taught and did research in the areas of evolution, development and data interoperability since 1997. She is the recipient of NIH and multiple NSF awards and authored over 80 research publications. She served as Division Director for the Division of Environmental Biology in the Directorate of Biological Sciences at the U.S. National Science Foundation (2015-2017). She was named an AAAS Fellow in 2004 for her fundamental studies in evolutionary and developmental biology.

Speaker Bios
Doug Johnson is the Associate Director of the Ohio Supercomputer Center. In his role at OSC, he manages the HPC Systems Operations and Engineering teams, which are responsible for the deployment, integration, and operation of the HPC systems. Johnson leads the center's efforts in the design and procurement of the computational, storage, data backup, and networking systems that comprise OSC's HPC systems. In his nearly three decades at OSC, he has led the transition to commodity Linux HPC clusters, and architected OSC's storage, data backup, and networking environments, and the implementation of OSC's Protected Data Service (PDS). He has also collaborated with many researcher groups to facilitate their use of HPC systems in their research.  Johnson earned his bachelor’s degree in physics from The Ohio State University.

Rob Guralnick is the Curator of Biodiversity Informatics at the Florida Museum of Natural History.  His work has involved developing standards for ecological and biodiversity data, building infrastructure and community capacity and developing collaborative frameworks.  His domain interests are firmly rooted in global change biology and conservation biogeography with a special interest in phenology and citizen science.  He is an elected fellow of the AAAS.

Jianwu Wang is a Professor of Data Science at the Department of Information Systems, University of Maryland, Baltimore County (UMBC). He leads the Center for Scalable Data and Computational Science (ScaleS) and the Big Data Analytics Lab (BDAL) at UMBC and co-leads the NSF HDR Institute for Harnessing Data and Model Revolution in the Polar Regions (iHARP). His research interests include Big Data Analytics, Causal AI, Earth Informatics and Distributed Computing. He has published 160+ papers with more than 4000 citations (h-index: 28). Since joining UMBC in 2015, he has received multiple external grants as PI (over $3.7M in total), Co-PI (over $13.7M in total) or Senior Personnel (over $20.3M in total) funded by ARL, NSF, NASA, DOE, State of Maryland, and Industry. He received Early-Career Faculty Excellence award from UMBC in 2019, NSF CAREER award in 2020 and Mid-Career Faculty Excellence award from UMBC in 2025.

After days of sharing ideas and exploring possibilities, this writing session gives you the space to act on those discoveries. Collaborate with fellow attendees to identify new research opportunities, forge potential partnerships, and turn conversations into concrete plans for future scientific exploration.