Welcome to AIVO News! Our newsletter features projects, events, and opportunities within the AI Institutes Virtual Organization community.
To subscribe, please complete the AIVO Newsletter Signup Form.
Contact us to suggest items for an upcoming newsletter issue or to provide feedback.
August 2024 Issue
Table of Contents:
Sign Up for SAIL by September 15 for the Best Pricing
Are you booked for SAIL? Register by September 15 for the best in-person rate.
Summit for AI Institutes Leadership
October 7–10, 2024
Pittsburgh Marriott City Center
Pittsburgh, PA
Both in-person and remote options are available. Visit the SAIL Registration page for details on registration options. Your SAIL conference registration includes admission to AI Institutes Expo Day at Carnegie Mellon University on October 10!
Register now through September 15 for $350.
Starting September 16, in-person registration is $500.
Remote participation registration is $150.
Don’t miss our esteemed keynote speakers!
U.S. Representative Summer Lee
Tom Mitchell, Founders University Professor, Carnegie Mellon University
Pat Yongpradit, Chief Academic Officer, Code.org
The SAIL 2024 website includes details on our keynote speakers and a list of sessions, as well as an FAQs page with answers to many common questions.
Questions or comments? Contact us at aivo@ucdavis.edu.
Sign Up for NSF OAC Cyberinfrastructure Office Hours
In the July issue of AIVO News, we announced that the National Science Foundation (NSF) Office of Advanced Cyberinfrastructure (OAC) Concierge Team is going to host Cyberinfrastructure (CI) office hours for the AI Institutes.
Thanks to all the AI Institute staff, researchers, and students involved in a center who provided their questions about NSF-funded CI resources as inputs for the content of the CI office hours!
If you’re interested in learning how you can make use of the advanced infrastructure resources provided by NSF, please attend their office hours, hosted by ACCESS. CI experts will answer your questions about what resources are available and how you can get access to them.
Complete the ACCESS-AIVO CI Office Hours sign-up form to join the inaugural office hours.
Connecting with the Public at AI Institutes Expo Day
AI Institutes Expo Day 2024 is our chance to showcase the forefront of AI research to the public!
Admission to Expo Day is included in the SAIL registration fee for AI Institute members.
October 10, 2024
Simmons Auditorium, Carnegie Mellon University
4765 Forbes Avenue
Pittsburgh, PA 15213
We’ll provide a peek into cutting-edge work in multiple fields and a glimpse into how AI is transforming and improving our world – including demonstrations of how AI is being used to tackle agricultural challenges, enhance education, and much more.
At Expo Day, attendees can:
Connect with 25 AI Institutes and partners and engage with their work
Watch interactive panel discussions with AI and industry experts
Network with 200+ attendees, researchers, and industry leaders
This event is affordable and open to the public. Visit our website to learn more about Expo Day: expo2024.aiinstitutes.org.
Please help us spread the word about Expo Day! Share our Expo Day flyer or post an Expo Day message for LinkedIn.
Momentum at the AI for Agriculture Summit
The AI for Agriculture Summit was held July 29–30, 2024 in Washington, D.C.
Convening AI institutes whose scope intersects in the AI for agriculture arena, experts from academia, industry, NGOs, and government agencies converged to envision how AI can tackle critical global challenges in agriculture.
There were many highlights! U.S. Senator Amy Klobuchar provided recorded remarks. USDA-National Institute of Food and Agriculture (NIFA) and NSF staff attended and addressed the summit audience.
- David Corman, Program Director, NSF
- Simon Liu, Administrator, Agricultural Research Services (ARS), USDA
- Christopher Alvares, Chief Data Officer, USDA
- Greg Hager, Assistant Director, Computer and Information Science and Engineering (CISE), NSF
- Jim Donlon, Program Director, NSF
Keynote addresses were provided by:
- Dionne Toombs, Associate Director for Programs, USDA-NIFA
- “AI for Agriculture: Advancing the Vision”
- Michael Littman, Division Director, Information and Intelligent Systems (CISE/IIS), NSF
- “NSF’s Role in the Future of AI”
- Janie Hipp, CEO/President, Native Agriculture Financial Services, longtime agricultural lawyer, and former General Counsel for USDA
- “Selected Legal Issues in an Evolving Landscape”
- Rachel Melnick, Division Director, Division of Global Climate Change, USDA-NIFA
- “Policy Levers for Climate and Environmental Justice for Agriculture”
Session panelists from both the public and private sectors discussed AI for agriculture:
- Saharah Moon Chapotin, Executive Director, Foundation for Food & Agriculture Research (FFAR)
- Leif Fixen, North America Agriculture Strategy Manager, The Nature Conservancy
- Joe Cornelius, CEO, Bill & Melinda Gates Agricultural Innovations
- Alton Thompson, Executive Director, Association of 1890 Research Directors
- Julian Reyes, Climate Adaptation Program Lead, Bureau of Land Management (BLM) and former Assistant Director for Climate Services, White House Office of Science and Technology Policy (OSTP)
- Walter Duflock, Senior Vice President, Innovation, Western Growers
- Erik Nicholson, Interim Executive Director, Semillero de Ideas
- Hendrik Hamann, Chief Science Officer for Climate and Sustainability, IBM Research
- Ranveer Chandra, Managing Director, Research for Industry, Microsoft
Summit Topics
- AI for Sensing/Perception
- AI for Agriculture Robotics
- AI for Crop Improvement
- AI for Decision-Making
- Data Ownership, Intellectual Property, and Labor Policy Considerations
- Environmental and Climate Policy
- Role of Public and Private Funding
White Paper
The goal of the summit is to develop a long-term technical vision and implementation strategy for leveraging agricultural AI advancements, starting with a white paper to share recommendations based on discussions held during the summit.
The white paper group met on July 31 to begin the organization and writing process for the white paper and to explore avenues for publication. See more photos from the summit.
Stay tuned for recordings of the summit sessions!
In the AI Institutes Community
Arboretum: A Large Multimodal Dataset Enabling AI for Biodiversity
Kudos to AIIRA! AIIRA recently released Arboretum, the largest publicly accessible #ai-ready dataset curated from the iNaturalist community science platform for advancing AI for biodiversity applications.
This domain-expert-vetted, image-language paired dataset, boasting 134.6 million samples, will be a valuable resource for building multimodal vision-language AI models for biodiversity assessment and agriculture research. They’re also releasing a suite of Contrastive Language-Image Pre-Training (CLIP) models – including Arbor-CLIP, trained using a subset of Arboretum (40 million captioned images) and report accuracy for zero-shot learning, and evaluations across life stages, rare species, confounding species, and various levels of the taxonomic hierarchy for a diverse set of species – including plants, insects, birds, reptiles, and fungi.
“We anticipate that due to its large scale and quality, Arboretum can spur the development of new families of AI foundation models that are specifically tailored for biodiversity and agriculture research. We have made some initial progress in this direction, but will welcome community feedback and input.
We hope that these new models can enable a variety of new digital tools for applications ranging from pest control, to crop monitoring, to worldwide biodiversity assessment and conservation.” — Chinmay Hegde, Faculty Researcher and Subaward Lead, AIIRA.
- Read the paper on Arboretum
- Visit the Arboretum website – It features the seven taxonomic classes in the Arboretum dataset, with example images of the top five most frequent species in each category, including their counts, common names, and scientific names, pictured above.
AI-CLIMATE and USDA’s Investment in GHG MMRV Improvement Efforts
USDA recently announced progress in efforts to improve the measurement, monitoring, reporting, and verification (MMRV) of greenhouse gas emissions (GHG)! One of its investments in this realm is the establishment of AI-CLIMATE.
“The Institute [AI-CLIMATE] — a joint effort between six Land-grant Universities — aims to advance foundational artificial intelligence by incorporating knowledge from agriculture and forestry sciences, and leveraging these unique, new methods to curb climate effects while lifting rural economies.”
Through a National Science Foundation (NSF) partnership, USDA is investing “$12 million to establish the AI Institute for Climate-Land Interactions, Mitigation, Adaptation, Tradeoffs and Economy (AI-CLIMATE) at the University of Minnesota.”
“We are grateful for the opportunity to investigate foundational and use-inspired AI for measurement, monitoring, reporting, and verification (MMRV) of greenhouse gas emissions (GHG) from agriculture and forests. For example, AI-CLIMATE is investigating novel knowledge guided machine learning models to improve the accuracy of GHG emission estimates for decision support Tools such as COMET Farm and COMET Planner.” Shashi Shekar, Director, AI-CLIMATE
AI Institutes at the International Conference on Machine Learning 2024
Multiple AI Institutes were in attendance – with students and postdocs presenting posters – at the International Conference on Machine Learning (ICML) 2024 conference!
This 41st international conference on machine learning, held at Messe Wien Exhibition Congress Center in Vienna, Austria, had more than 9,000 participants. The conference included:
9 Expo Panels
12 Tutorials
6 Invited Talks
Over 2,500 poster presentations
30 different workshops
Visit the ICML post on the AIVO website to see more pictures from the event.
Happenings in SIGs
Updates on Team Science Education Training for AIVO PMs
In the July issue of AIVO News, we announced that Team Science Education is launching within the Project Management SIG. The launch includes several engagements with Divergent Science. There will be sessions this summer and fall, culminating with a SIG session at SAIL 2024.
On July 16, Team Science had a call featuring a group of AIVO Project Managers from the AI Institutes, as well as the personnel from Divergent Science who are coordinating the PM training, Hannah Love and Ellen Fisher.
On August 20, Team Science had its official kickoff call. PMs offered up topics that they want to see covered at the in-person engagement at SAIL, which will be on Monday, October 7.
They identified some key core values for the group as they pursue the training/white paper, which largely revolve around values such as openness, collaboration, respect, accountability, and learning. The PMs discussed a range of topics for future office hour sessions related to project management, including effective communication, team dynamics, motivation, and career development. They also discussed challenges specific to AI research and interdisciplinary collaboration. Potential white paper topics centered around the value of project managers, strategies for motivating teams, and best practices for managing diverse and complex projects.
Five to six hour-long office hour sessions are being scheduled between now and SAIL, and online modules have been disseminated to the team to review before each office hours session.
In the coming months, Divergent Science will continue to refine their goals for the in-person workshop at SAIL.
New on the AIVO Website: SIG Pages
Read up on all five special interest groups (SIGs). Contact the AIVO team to add resources to your SIG page.
Ethics and Trustworthiness SIG
