Lily Xu
PhD student at Harvard
I am a PhD student in computer science at Harvard University, advised by Prof. Milind Tambe. My research focuses on developing AI methods across machine learning, sequential decision-making, and optimization for planetary health, particularly to address environmental challenges such as biodiversity conservation.
I co-organize the Mechanism Design for Social Good (MD4SG) research initiative, which advances computational techniques to improve access to opportunity for historically marginalized communities. Additionally, I serve as AI Lead for the SMART Partnership, where I help build computational and research solutions for effective conservation management.
News
I am on the 2023–2024 academic job market.
For more about me, see my research statement and teaching statement.
- 2023: I'm honored to have been named a Rising Star in Management Science & Engineering (at Stanford), Data Science (at UChicago), and EECS (at Georgia Tech), and to be speaking at the Cornell ORIE Young Researchers Workshop!
- November 2023: I'll be in London to attend part of the UK Global Summit on AI Safety in Bletchley Park, in my role as AI Lead for the SMART Partnership.
- October 2023: I'll be at EAAMO 2023, co-chairing social events and co-leading the tutorial on "Towards effective, deployed, and thoughtful AI for Social Impact". Join us in Boston to discuss bridging research and practice!
- October 2023: I'll be speaking at the Berkeley Institute for Data Science.
- July 2023: I'll be attending the International Congress for Conservation Biology (ICCB) in Kigali, Rwanda! I'll present in a symposium on protected area management, hosted by the SMART Partnership.
- July 2023: Speaking at the Harvard Kennedy School Executive Education class on Leading in Artificial Intelligence: Exploring Technology and Policy.
- Spring 2023: I'll be giving guest lectures at Harvard, MIT, and Cornell, and *in-person* talks at Berkeley and Brown!
- April 2023: I'll be giving a talk at the WWF Fuller Seminar Series. (Watch the recording on Vimeo)
- October 2022: Organizing the workshop on AI-Assisted Decision Making for Conservation with Esther Rolf and Milind Tambe. I'm very excited to bring ecologists, computer scientists, and conservation practitioners all in a room together!
- July 2022: I'm excited to moderate a panel on "Building Bridges Between AI Research and Policy" at the Hertz workshop! Panelists are Julie Owono, Jessica Fjeld, and Jacob Steinhardt.
Research interests
- Sequential decision making: multi-armed bandits, reinforcement learning, robust planning, game theory
- Machine learning: causal inference; data science; learning in sparse, noisy settings
- AI for social impact: AI for conservation, bridging research and practice, community building
Working papers
Reinforcement learning with combinatorial actions for coupled restless bandits
paper bib@inproceedings{xu2023reinforcement, title={Reinforcement learning with combinatorial actions for coupled restless bandits}, author={Xu, Lily and Wilder, Bryan and Khalil, Elias B. and Tambe, Milind}, booktitle={arXiv}, year={2023}, }
under review
Lily Xu, Bryan Wilder, Elias B. Khalil, Milind Tambe
Reflections from the Workshop on AI-Assisted Decision Making for Conservation paperbib
@inproceedings{xu2023reflections, title={Reflections from the Workshop on {AI}-Assisted Decision Making for Conservation}, author={Xu, Lily and Rolf, Esther and Beery, Sara and Bennett, Joseph R. and Berger-Wolf, Tanya and Birch, Tanya and Bondi-Kelly, Elizabeth and Brashares, Justin and Chapman, Melissa and Corso, Anthony and Davies, Andrew and Garg, Nikhil and Gaylard, Angela and Heilmayr, Robert and Kerner, Hannah and Klemmer, Konstantin and Kumar, Vipin and Mackey, Lester and Monteleoni, Claire and Moorcroft, Paul and Palmer, Jonathan and Perrault, Andrew and Thau, David and Tambe, Milind}, booktitle={arXiv}, year={2023}, }
white paper
Lily Xu*, Esther Rolf*, Sara Beery, Joseph R. Bennett, Tanya Berger-Wolf, Tanya Birch, Elizabeth Bondi-Kelly, Justin Brashares, Melissa Chapman, Anthony Corso, Andrew Davies, Nikhil Garg, Angela Gaylard, Robert Heilmayr, Hannah Kerner, Konstantin Klemmer, Vipin Kumar, Lester Mackey, Claire Monteleoni, Paul Moorcroft, Jonathan Palmer, Andrew Perrault, David Thau, Milind Tambe
Artificial Replay: A Meta-Algorithm for Harnessing Historical Data in Bandits paper code bib
@inproceedings{banerjee2023artificial, title={Artificial Replay: A Meta-Algorithm for Harnessing Historical Data in Bandits}, author={Banerjee, Siddhartha and Sinclair, Sean R. and Tambe, Milind and Xu, Lily and Yu, Christina Lee}, booktitle={arXiv}, year={2023}, }
under review
(alphabetical) Siddhartha Banerjee, Sean R. Sinclair, Milind Tambe, Lily Xu, Christina Lee Yu
Ranger patrols deter poaching: first causal insights for improving protected area management bib
@article{guo2023ranger, title={Ranger patrols deter poaching: first causal insights for improving protected area management}, author={Guo, Rachel and Xu, Lily and Perrault, Andrew and Miratrix, Luke W. and Plumptre, Andrew J. and Mabonga, Joshua and Kitimbo, Herbert and Wanyama, Fredrick and Nampindo, Simon and Tambe, Milind and Davies, Andrew B.}, journal={under review}, year={2023}, }
under review
Rachel Guo*, Lily Xu*, Andrew Perrault, Luke W. Miratrix, Andrew J. Plumptre, Joshua Mabonga, Herbert Kitimbo, Fredrick Wanyama, Simon Nampindo, Milind Tambe, Andrew B. Davies
Publications
* indicates equal contribution
Optimistic Whittle Index Policy: Online Learning for Restless Bandits paper code bib
@inproceedings{wang2023online, title={Optimistic {W}hittle Index Policy: Online Learning for Restless Bandits}, author={Wang, Kai and Xu, Lily and Taneja, Aparna and Tambe, Milind}, booktitle={Proc.~Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI)}, year={2023}, }
AAAI 2023 37th AAAI Conference on Artificial Intelligence
Kai Wang*, Lily Xu*, Aparna Taneja, Milind Tambe
Robust Planning over Restless Groups: Engagement Interventions for a Large-Scale Maternal Telehealth Program paper talk bib
@inproceedings{killian2023groups, title={Robust Planning over Restless Groups: Engagement Interventions for a Large-Scale Maternal Telehealth Program}, author={Killian, Jackson A. and Biswas, Arpita and Xu, Lily and Verma, Shresth and Nair, Vineet and Taneja, Aparna and Hegde, Aparna and Madhiwalla, Neha and Rodriguez Diaz, Paula and Johnson-Yu, Sonja and Tambe, Milind}, booktitle={Proc.~Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI)}, year={2023}, }
AAAI 2023 37th AAAI Conference on Artificial Intelligence
Jackson A. Killian*, Arpita Biswas*, Lily Xu*, Shresth Verma*, Vineet Nair, Aparna Taneja, Aparna Hegde, Neha Madhiwalla, Paula Rodriguez Diaz, Sonja Johnson-Yu, Milind Tambe
Flexible Budgets in Restless Bandits: A Proximal Primal-Dual Algorithm for Efficient Budget Allocation paper bib
@inproceedings{rodriguez2023online, title={Flexible Budgets in Restless Bandits: A Proximal Primal-Dual Algorithm for Efficient Budget Allocation}, author={Rodriguez Diaz, Paula and Killian, Jackson A. and Xu, Lily and Taneja, Aparna and Suggala, Arun Sai and Tambe, Milind}, booktitle={Proc.~Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI)}, year={2023}, }
AAAI 2023 37th AAAI Conference on Artificial Intelligence
Paula Rodriguez Diaz, Jackson A. Killian, Lily Xu, Aparna Taneja, Arun Sai Suggala, Milind Tambe
Bridging adaptive management and reinforcement learning for more robust decisions paper preprint bib
@article{chapman2023bridging, title={Bridging adaptive management and reinforcement learning for more robust decisions}, author={Chapman, Melissa and Xu, Lily and Lapeyrolerie, Marcus and Boettiger, Carl}, journal={Philosophical Transactions of the Royal Society B}, volume={378}, number={1881}, pages={20220195}, year={2023}, publisher={The Royal Society}, }
Phil Trans B 2023 Philosophical Transactions of the Royal Society B
Melissa Chapman, Lily Xu, Marcus Lapeyrolerie, Carl Boettiger
Environment, Society, and Machine Learning (book chapter) bib
@incollection{scoville2023environment, author = {Scoville, Caleb and Faxon, Hilary and Chapman, Melissa and Fried, Samantha Jo and Xu, Lily and Boettiger, Carl and Reed, J. Michael and Lapeyrolerie, Marcus and Van Scoyoc, Amy and Amironesei, Razvan}, title = {Environment, Society, and Machine Learning}, editor = {Borch, Christian and Pardo-Guerra, Juan Pablo}, booktitle = {Oxford Handbook of the Sociology of Machine Learning}, publisher = {Oxford University Press}, year = {2023}, }
Oxford University Press 2023 Oxford Handbook of the Sociology of Machine Learning, forthcoming
Caleb Scoville, Hilary Faxon, Melissa Chapman, Samantha Jo Fried, Lily Xu, Carl Boettiger, J. Michael Reed, Marcus Lapeyrolerie, Amy Van Scoyoc, and Razvan Amironesei
Restless and Uncertain: Robust Policies for Restless Bandits via Deep Multi-Agent Reinforcement Learning paper code bib
@inproceedings{killian2022restless, title={Restless and Uncertain: Robust Policies for Restless Bandits via Deep Multi-Agent Reinforcement Learning}, author={Killian, Jackson A. and Xu, Lily and Biswas, Arpita and Tambe, Milind}, booktitle={Proc.~38th Conference on Uncertainty in Artificial Intelligence (UAI)}, year={2022}, }
UAI 2022 38th Conference on Uncertainty in Artificial Intelligence
Jackson A. Killian, Lily Xu, Arpita Biswas, Milind Tambe
Ranked Prioritization of Groups in Combinatorial Bandit Allocation paper code talk bib
@inproceedings{xu2022ranked, title={Ranked Prioritization of Groups in Combinatorial Bandit Allocation}, author={Xu, Lily and Biswas, Arpita and Fang, Fei and Tambe, Milind}, booktitle={Proc.~31st International Joint Conference on Artificial Intelligence (IJCAI)}, year={2022}, }
IJCAI 2022 31st International Joint Conference on Artificial Intelligence
Lily Xu, Arpita Biswas, Fei Fang, Milind Tambe
Coordinating Followers to Reach Better Equilibria: End-to-End Gradient Descent for Stackelberg Games paper code bib
@inproceedings{wang2022coordinating, title={Coordinating Followers to Reach Better Equilibria: End-to-End Gradient Descent for {S}tackelberg Games}, author={Wang, Kai and Xu, Lily and Perrault, Andrew and Reiter, Michael K. and Tambe, Milind}, booktitle={Proc.~36th AAAI Conference on Artificial Intelligence (AAAI)}, year={2022}, }
AAAI 2022 36th AAAI Conference on Artificial Intelligence
Kai Wang, Lily Xu, Andrew Perrault, Michael K. Reiter, Milind Tambe
Robust Reinforcement Learning Under Minimax Regret for Green Security paper code talk bib
@inproceedings{xu2021robust, title={Robust Reinforcement Learning Under Minimax Regret for Green Security}, author={Xu, Lily and Perrault, Andrew and Fang, Fei and Chen, Haipeng and Tambe, Milind}, booktitle={Proc.~37th Conference on Uncertainty in Artificial Intelligence (UAI)}, year={2021}, }
UAI 2021 37th Conference on Uncertainty in Artificial Intelligence
Lily Xu, Andrew Perrault, Fei Fang, Haipeng Chen, Milind Tambe
★ Oral Presentation
Envisioning Communities: A Participatory Approach Towards AI for Social Good paper talk bib
@inproceedings{bondi2021envisioning, title={Envisioning Communities: A Participatory Approach Towards {AI} for Social Good}, author={Bondi, Elizabeth and Xu, Lily and Acosta-Navas, Diana and Killian, Jackson A.}, booktitle={Proc.~4th AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society (AIES)}, year={2021}, }
AIES 2021 4th AAAI/ACM Conference on AI, Ethics, and Society
Elizabeth Bondi*, Lily Xu*, Diana Acosta-Navas, Jackson A. Killian
Dual-Mandate Patrols: Multi-Armed Bandits for Green Security paper code talk bib
@inproceedings{xu2021dual, title={Dual-Mandate Patrols: Multi-Armed Bandits for Green Security}, author={Xu, Lily and Bondi, Elizabeth and Fang, Fei and Perrault, Andrew and Wang, Kai and Tambe, Milind}, booktitle={Proc.~35th AAAI Conference on Artificial Intelligence (AAAI)}, year={2021}, }
AAAI 2021 35th AAAI Conference on Artificial Intelligence
Lily Xu, Elizabeth Bondi, Fei Fang, Andrew Perrault, Kai Wang, Milind Tambe
★ Best Paper Award Runner Up
Stay Ahead of Poachers: Illegal Wildlife Poaching Prediction and Patrol Planning Under Uncertainty with Field Test Evaluations paper code talk bib
@inproceedings{xu2020stay, title={Stay Ahead of Poachers: Illegal Wildlife Poaching Prediction and Patrol Planning Under Uncertainty with Field Test Evaluations}, author={Xu, Lily and Gholami, Shahrzad and Mc Carthy, Sara and Dilkina, Bistra and Plumptre, Andrew and Tambe, Milind and Singh, Rohit and Nsubuga, Mustapha and Mabonga, Joshua and Driciru, Margaret and Wanyama, Fred and Rwetsiba, Aggrey and Okello, Tom and Enyel, Eric}, booktitle={Proc.~IEEE 36th International Conference on Data Engineering (ICDE)}, year={2020}, }
ICDE 2020 36th IEEE International Conference on Data Engineering
Lily Xu*, Shahrzad Gholami*, Sara Mc Carthy, Bistra Dilkina, Andrew Plumptre, Milind Tambe, Rohit Singh, Mustapha Nsubuga, Joshua Mabonga, Margaret Driciru, Fred Wanyama, Aggrey Rwetsiba, Tom Okello, Eric Enyel
Workshop Papers
Learning, Optimization, and Planning Under Uncertainty for Wildlife Conservation paper talk
Preprint, for the INFORMS Doing Good with Good OR Competition 2021
Lily Xu
Enhancing Poaching Predictions for Under-Resourced Wildlife Conservation Parks Using Remote Sensing Imagery paper bib
@inproceedings{guo2020enhancing, title={Enhancing Poaching Predictions for Under-Resourced Wildlife Conservation Parks Using Remote Sensing Imagery}, author={Guo, Rachel and Xu, Lily and Plumptre, Andrew and Cronin, Drew and Okeke, Francis and Tambe, Milind}, booktitle={NeurIPS Workshop on Machine Learning for the Developing World}, year={2020} }
ML4D 2020 NeurIPS Workshop on Machine Learning for Development
Rachel Guo, Lily Xu, Andrew Plumptre, Drew Cronin, Francis Okeke, Milind Tambe
Selected Media
- HBS case study. SMART: AI and Machine Learning for Wildlife Conservation
with supporting episode on HBS Cold Call (23 min podcast) - Business Insider. Scientists Use Artificial Intelligence Tools to Help Save Wildlife
- INFORMS Resoundingly Human podcast. Giving wildlife conservationists some helping PAWS
[Spotify] [Apple Podcasts] (37 min) - Voice of America News. Wildlife Rangers Use AI to Predict Poachers’ Next Moves (2 min video)
- Harvard SEAS. Computer Conservation
- AIhub. Interview with Lily Xu: applying machine learning to the prevention of illegal wildlife poaching
- Rethinking Economics NL podcast, discussing Mechanism Design for Social Good with Jessie Finocchiaro, Sera Linardi, Faidra Monachou, and Ana-Andreea Stoica
[YouTube] [Spotify] [Apple Podcasts] (1 hour) - Mongabay. Where to patrol next: ‘Netflix’ of ranger AI serves up poaching predictions
Talks and recordings
- WWF Fuller Seminar on Causality for Conservation, April 2023 (59 min)
- Berkeley Multi-Agent Learning Seminar on Minimax Regret Optimization: Robust Planning for Conservation and Maternal Health, May 2023 (50 min)
- Penn State CSRAI Young Achievers Symposium, March 2023 (1 h 6 min)
- Analytics for a Better World (ABW) research meetup, September 2022 (56 min)
- Future of AI for Social Impact panel, April 2022 (1 h 27 min)
- WILDLABS Tech Tutors. How do I use AI to fight wildlife crime? September 2021 (1 hour 16 min video)
Contact
[first]_[last] at g.harvard.edu
Harvard SEC Room 2.106
150 Western Avenue
Allston, MA 02134