AI-Assisted Decision-Making for Conservation

The Harvard CRCS workshop on AI-Assisted Decision-Making for Conservation aims to identify open research questions in resource-allocation, planning, and interventions for biodiversity conservation. In particular, we seek to pinpoint conservation challenges that not only require AI solutions, but require novel methodological advances. This workshop will bring together thought leaders across academia, industry, government, and NGOs with expertise in computation, conservation, and deploying research.

Much existing work at the intersection of AI and conservation uses computer vision for biodiversity monitoring, such as supervised learning approaches to classify animals or land cover. This workshop aims to look ahead at ways to leverage and develop AI methodology for planning and policy-making, with the goal of helping conservation managers make decisions.

Schedule

PDF of slides available for each talk.

Participants: for detailed schedule and talk information, please refer to the Google Doc shared via email by Lily and Esther.

Day 1: Challenges and Opportunities in Conservation
Thursday, October 20

8:30am	Breakfast and check-in
9am	Introductions Lily Xu, Esther Rolf, and Milind Tambe Workshop overview
9:30am	Session 1: Protected area management Angela Gaylard, African Parks: Adaptive management of complex systems Justin Brashares, Protected area and private land conservation + community engagement Jonathan Palmer, Protected Areas and Conservation Technology Millie Chapman, Spatial planning to support global biodiversity targets: navigating governance structures, data disparities, & dynamic ecosystems Lily Xu, Learning and planning under uncertainty for conservation
11:30am	Lunch
12:30pm	Session 2: Adaptive management Paul Moorcroft, Predictive models of animal movement Joe Bennett, Decision-support tools for biodiversity conservation: progress and challenges Andrew Davies, Animal-Landscape Interactions on a Changing Planet Anthony Corso, Large-scale planning for resource management problems Elizabeth Bondi-Kelly, Human-AI collaboration for conservation
2:20pm	Break
2:40pm	Session 3: AI methods for conservation Tanya Berger-Wolf, Community engaged trustworthy AI for conservation Dave Thau, Artificial intelligence and decision making at WWF Lester Mackey, Improving subseasonal forecasting with machine learning Esther Rolf, MOSAIKS: General and accessible machine learning with satellite imagery Sara Beery, Open Challenges in Generalizable Computer Vision for Conservation
4:30pm	Recap, discussion, and goal setting
5pm	Break
6pm	Dinner at Forage in Harvard Square (indoor) 5 Craigie Circle, Cambridge, MA 02138

Day 2: Methods and Synthesis
Friday, October 21

8:30am	Breakfast
9:00am	Session 4: Remote sensing and earth observation Vipin Kumar, ML and Earth Observation Data: Challenges and Opportunities for Conservation Claire Monteleoni, AI Research for Climate Change and Environmental Sustainability Tanya Birch, Dynamic World and Wildlife Insights Hannah Kerner, AI for the Digital Planet Konstantin Klemmer, Tackling spatial residual autocorrelation (RSA) in ecological models and other geographic applications
10:50am	Break
11:20am	Session 5: Causality and economics for conservation Robert Heilmayr, Conservation Impact Evaluation Using Remotely Sensed Data Nikhil Garg, Auditing and Designing for Equity in Government Service Allocation Andrew Perrault, Sequential decision-making in complex partially observable environments
12:30pm	Lunch
1:30pm	Collaborative writing
2:15pm	Panel on partnerships: the role of academia Tanya Berger-Wolf, Justin Brashares, Claire Monteleoni, Jonathan Palmer, Milind Tambe
3pm	Collaborative writing, part 2
4pm	Break
5pm	Happy hour at Wusong Road in Harvard Square (indoor) 112 Mt Auburn St. Cambridge MA 02138

Participants

Andrew Davies (Harvard)
Andrew Perrault (Ohio State)
Angela Gaylard (African Parks)
Anthony Corso (Stanford)
Claire Monteleoni (CU Boulder)
Dave Thau (WWF)
Elizabeth Bondi-Kelly (MIT / Michigan)
Hannah Kerner (Arizona State)
Joe Bennett (Carleton)
Jonathan Palmer (WCS)
Justin Brashares (Berkeley)
Konstantin Klemmer (Microsoft Research)
Lester Mackey (Microsoft Research)
Melissa (Millie) Chapman (Berkeley)
Nikhil Garg (Cornell Tech)
Paul Moorcroft (Harvard)
Robert Heilmayr (UCSB Bren)
Sara Beery (Google / MIT)
Tanya Berger-Wolf (Ohio State)
Tanya Birch (Google Geo)
Vipin Kumar (Minnesota)

Organizers

Lily Xu Harvard	Esther Rolf Harvard	Milind Tambe Harvard / Google

Lily Xu
Harvard

Esther Rolf
Harvard

Milind Tambe
Harvard / Google

Please don't hesitate to reach out to us with any questions, ideas, or concerns!

Workshop logistics

Dates: October 20–21, 2022

Location: Harvard SEC, Room 4.307 + 4.308
Science and Engineering Complex
150 Western Ave, Boston, MA 02134

The SEC is located on Harvard's Allston campus, next to Harvard Business School.

Parking is available at the SEC, but please let us know if you will need a spot.

Hotel booking

We have secured discounted hotel rates for workshop attendees at the Hyatt Regency Boston / Cambridge. These will be the most affordable rates by far. Please see our registration email for details on how to book.

COVID policy

Vaccinations: Harvard University currently has a vaccination requirement for all University community members. All visitors to campus should expect to provide proof of full COVID-19 vaccination and booster or proof of a negative COVID-19 PCR or rapid antigen test taken within one calendar day of the event(s) on campus. More details on Harvard COVID policies.

Masking: All persons are encouraged wear a mask of any type for their own protection, or to protect someone else, if they choose to do so. We will offer university-provided surgical masks for all attendees.

Outdoor dining: We have an outdoor terrace reserved for the entire day, including breakfast, lunch, and coffee breaks. Happy hour will be outdoors; dinner will be indoors due to challenges of finding a sufficiently large outdoor venue.