The ICDS Symposium will include many activities designed to share knowledge or encourage interdisciplinary collaboration.

The following tentative agenda is subject to change. All times listed are in ET.

Please note: all sessions will be recorded.

Wednesday, October 21

All activities to take place virtually via Zoom. You must register to have access to the links.

Session 1: 8:00 – 9:30 a.m.

8:00 a.m.: Introductory Remarks:
Jenni Evans, Director, Institute for Computational and Data Sciences

8:15 a.m.: Keynote Presentation: “ZettaScale Computing on Exascale Platforms!”
Presented by Shantenu Jha, Chair, Computation and Data Driven Discovery (C3D) Department at Brookhaven National Laboratory, and Professor of Computer Engineering at Rutgers University  

Session 2: Noon – 1:30 p.m. Panel Discussions (descriptions listed below)

Panel 1: Big Data, Agriculture & Food Supply

Panel 2: Artificial Intelligence and Machine Learning in Manufacturing

Session 3a: 3:00 – 3:30 p.m.: Sponsor Showcase – Product Briefings

Session 3: 3:30 – 5:00 p.m.: (ICDS Connects – Industry event) 

State of the Industry talks – 30 minutes 

Faculty Lighting Talks – 2-3 min each (total of 60 min) 

Thursday, October 22

All activities to take place virtually via Zoom. You must register to have access to the links.

Session 1: 8:00 – 9:30 a.m.

8:00 a.m.: Introductory Remarks:
Jenni Evans, Director, Institute for Computational and Data Sciences
8:15 a.m.: Keynote Presentation: “The Landscape of Data Science: Basic Research to Impact.”
Presented by Chaitan Baru, Senior Science Advisor, Convergence Accelerator, Office of the Director, National Science Foundation

Session 2: Noon – 1:30 p.m.: Panel Discussions (descriptions listed below)

Panel 1: Social Engineering with Data: Disinformation & Destabilization of Geo-Political Order

Panel 2: Data & Genetics/DNA: Value, Ethics, and Risks

Keynote Descriptions

ZettaScale Computing on Exascale Platforms, presented by Shantenu Jha

Abstract: We outline the vision of  “Learning Everywhere,” which captures the impact of learning methods coupled to traditional HPC methods. We: (i) discuss effective performance improvements for traditional HPC simulations that learning (MLforHPC) provides; (ii) provide a taxonomy of the modes by which MLforHPC can impact computational science, including scenarios: MLinHPC, MLoutHPC and MLaroundHPC; and (iii) identify and survey recent problems that benefit from MLforHPC. We will also outline software systems developed for ML driven simulations and discuss how learning methods and HPC simulations are being integrated. We identify a spectrum of challenges and requirements that MLforHPC presents for both new cyberinfrastructure and application developments.

The Landscape of Data Science: Basic Research to Impact, presented by Chaitan Baru

Abstract: Over the past three years, via its Harnessing the Data Revolution Big Idea (aka HDR), and other related programs, the National Science Foundation has launched a series of multidisciplinary programs covering foundations, systems, applications, and education in Data Science. For example, the Transdisciplinary Research In Principles Of Data Science (TRIPODS) program explores the foundations of data science at the nexus of computer science, statistics, and mathematics. The TRIPODS+X program explores how the data challenges and concepts from various science domains (the “X”) might interact with and influence foundational issues. The NSF HDR Institutes program seeks to establish center-scale activities in data science encompassing aspects of theory, systems, and applications of data science methods across various disciplines and applications. The Data Science Corps program supports the development of experiential learning curricula in undergraduate data science education. The NSF Convergence Accelerator is a new, unique program to support use-inspired convergent research characterized by deep multidisciplinary collaborations and partnerships among academia, industry, government, non-profit and other sectors, with the goal of accelerating ideas from research into practice. In 2019, its first pilot year, the NSF Convergence Accelerator is supporting projects in two tracks that involve data science: the Open Knowledge Network and AI and Future Work.

The range of new, data science-related programs and the variety of programmatic approaches being taken at NSF reflects the excitement and experimentation underway in academia as well, where a variety of new data science paths are being explored…almost as many as there are universities!

In this discussion, we will explore the landscape of research programs and activities in data science; examine what makes data science new and different from programs we have seen thus far; and consider future directions.

Panel Descriptions

Wednesday Panel 1: Big Data, Agriculture & Food Supply

Organized by Asad Azemi, Associate Professor of Engineering, Penn State Brandywine


  • David Hughes, Associate Professor of Entomology & Biology, Penn State University
  • Paul Esker, Assistant Professor, Epidemiology and Field Crop Pathology, Penn State University
  • Long He, Assistant Professor, Agricultural and Biological Engineering, Penn State University

Data that is unstructured or time sensitive or simply very large cannot be processed by relational database engines. Such datasets in agriculture often include numerous weather and soil measurements as well as corresponding plant or animal performance assessments under multiple management regimes over multiple years. This type of data requires a different processing approach called big data, which uses massive parallelism on readily-available hardware. 

Predictive models derived from big data can help to identify best management practices for getting the best crop and livestock performance under various environmental conditions, and help to make decisions that will tackle inefficiencies in planting, harvesting, water use and energy, and increase yields and deliver safe, nutritious food to communities around the world. 

Join us as interdisciplinary panelists address how big data and AI are improving food and agriculture from farm to table. 

Wednesday Panel 2: Artificial Intelligence and Machine Learning in Manufacturing

Organized by Soundar Kumara, Allen E. Pearce and Allen M. Pearce Professor of Industrial Engineering


  • Josh Siegel, Assistant Professor of Computer Science and Engineering, Michigan State University
  • Paul Witherell, Mechanical Engineer, Systems Integration Division, National Institute of Standards and Technology (NIST)

This panel aims to facilitate a discussion on exploring the application of artificial intelligence (AI) and machine learning (ML) in the future of manufacturing systems. Manufacturing systems have evolved from the early computer integrated manufacturing to the current Department of Energy’s Smart Manufacturing. In recent years, the proliferation of sensors and data analytics have resulted in research ideas and implemented systems that are termed, “smart.” AI and ML have evolved in the last decade to be the most important technologies to enhance every aspect of human life. Given the speed with which AI and ML have evolved in the last five years, disciplinary areas such as manufacturing have not moved at a commensurate speed. In this context, it is necessary to look at manufacturing from the perspective of AI and ML driven society of the future and prepare to lay the fundamental tenants of research, development and education. Manufacturing needs to take not only process and system level aspects but also socio, economic and political changes that are influencing the way we live, work and collaborate. The objective of this panel is to bring together researchers and practitioners to discuss and generate a roadmap for AI and ML driven manufacturing research, development and education.

Thursday Panel 1: Social Engineering with Data: Disinformation & Destabilization of Geo-Political Order

Organized by Anne Toomey McKenna, Distinguished Scholar of Cyber Law & Policy, Penn State Dickinson Law, and Co-Hire, Institute for Computational and Data Sciences


  • Kate Hammerberg, Research Director, Analytics and Research Team, Global Engagement Center, U.S. Department of State
  • Anthony C. Robinson, Associate Professor, Director, Online Geospatial Education, Assistant Director, GeoVISTA center, Department of Geography, Penn State
  • Kevin Munger, Penn State, Assistant Professor of Political Science and Social Data Analytics
  • Maria D. Molina, PhD Candidate, Penn State Bellisario College of Communications

The U.S. and other nations are the testing and proving grounds for large-scale social engineering with data. These efforts arguably are transforming existing geo-political order and threaten the foundations of democracy, including fair and accurate elections. Harnessing vast quantities of consumer data, social engineering is the online manipulation of citizens via disinformation and targeted behavioral messaging (using data) on social media platforms (information eco-systems). These intentional efforts by nation states and private actors include:

  • manipulation of social groups and vulnerable populations
  • engineering election results
  • erosion of confidence in the electoral process
  • undermining democracy
  • altering geo-political order

Join us as four interdisciplinary panelists address how researchers, citizens, and governments are using AI, cyber measures and other security technologies, and the law, to investigate and identify disinformation, secure electoral systems, mitigate destabilization, educate the public about social engineering with data, and create policies that combat malicious social engineering.

Thursday Panel 2: Data & Genetics/DNA: Value, Ethics, and Risks

Organized and moderated by Aleksandra (Sesa) Slavkovic, Professor; Associate Dean for Graduate Education, Eberly College of Science


  • Daniel Kifer, Associate Professor, Computer Science and Engineering
  • Elizabeth McGraw, Director of the Center for Infectious Disease Dynamics; Professor and Huck Scholar in Entomology
  • Nilam Ram, Professor, Human Development and Family Studies, and Psychology
  • Daniel Susser, Assistant Professor of Information Sciences and Technology and Philosophy, Research Associate, Rock Ethics Institutes


The data deluge brought forth a great deal of discussion about the four V’s of big data: Volume, Variety, Velocity, and Veracity. But intertwined with these aspects are:

  • VALUE — to what extent are the data needed and insights gained from analyses via statistical, ML, or AI models and systems impactful?
  • ETHICS — what are the normative issues in generating, analyzing and disseminating data? 
  • RISKS — how do we think about and define risks in the scientific enterprise that relies on data?

While these issues most naturally arise within contexts that deal with human data, any scientific discipline should consider these dimensions to enable sounds scientific progress and decision making. We will hear perspectives on these three topics, including related opportunities and challenges, from our interdisciplinary panel of experts. We will discuss how the human screenome project aims to capture our digital lives, and investigate its relation to genomics. We will also discuss new directions in privacy, transparency and reproducibility using census data.  More broadly, we will examine how technology aids in online manipulation and is impacting our autonomy, and how life sciences research spanning the bench, modeling and the field impacts our understanding of human disease and public health.

ICDS Symposium Sponsors