About

Sanmay Das is a professor of computer science and associate director of AI for social impact at the Sanghani Center for Artificial Intelligence and Data Analytics at the Virginia Tech Institute for Advanced Computing in Alexandria, Virginia. He is a AAAI Fellow and an ACM Distinguished Member. Das has served as past chair of ACM SIGAI, a member of the DARPA ISAT Study Group, and an emeritus member of the board of directors of the International Foundation for Autonomous Agents and Multiagent Systems. He also serves as an arXiv moderator and has held numerous editorial roles, including associate editor for the ACM Transactions on Economics and Computation, the Journal of Artificial Intelligence Research, and Autonomous Agents and Multiagent Systems. His leadership roles include program co-chair and general co-chair of AAMAS and the AAAI/ACM Conference on AI, Ethics, and Society, as well as associate program chair for IJCAI. Das has received awards for research, teaching, and service, such as a National Science Foundation CAREER Award, the Department Chair Award for Outstanding Teaching at Washington University, and the Outstanding Service Award from the Computer Science Department at George Mason University. His research interests encompass artificial intelligence, machine learning, computational social science, AI for public services and resources, and AI-mediated human systems. He holds a Ph.D. in Computer Science and an M.S. in Electrical Engineering and Computer Science from the Massachusetts Institute of Technology, as well as an A.B. in Computer Science from Harvard University.

Research topics

• Computer Science
• Sociology
• Economics
• Medicine
• Political Science
• Business
• Mathematics
• Virology
• Biology
• Environmental health
• Economic growth
• Mathematical economics
• Engineering
• Public administration
• Demography
• Political economy
• Law

Selected publications

• Federated Parameter-Efficient Adaptation for Interference Mitigation at the Wireless Edge

  arXiv (Cornell University) · 2026-04-17

  preprintOpen accessSenior author

  Dense wireless deployments face co-channel interference from heterogeneous sources that vary across base stations (gNBs in 5G). While centralized DNN-based approaches to interference mitigation have shown strong performance, deploying and adapting these models across distributed gNBs via federated learning (FL) requires transmitting full model updates each round, resulting in a cost that scales poorly with network density. Parameter-efficient fine-tuning (PEFT) reduces this burden by training and communicating only a small fraction of parameters. While traditionally applied to large foundation models, we adapt Low-Rank Adaptation (LoRA) to temporal convolutional neural network architectures for interference suppression, placing low-rank adapters on the dilated convolutional layers. This placement enables LoRA to learn local interference-specific temporal patterns, while the frozen backbone retains the shared signal extraction capability. These lightweight adapters (5.1\% of backbone parameters) are federated via FedAvg, reducing per-round communication by up to 20$\times$ compared to federating full model updates. We evaluate various PEFT strategies across simulated distributed gNBs with non-IID interference environments. Results show that local LoRA achieves 12.8\% average BER improvement over the frozen backbone, while Fed-LoRA achieves comparable performance (12.6\%). Fed-LoRA outperforms local adaptation on data-starved nodes where federated knowledge transfer compensates for limited samples, all while avoiding the catastrophic degradation observed with full-model FedAvg under heterogeneous conditions.

  PublisherDOI

• Federated Parameter-Efficient Adaptation for Interference Mitigation at the Wireless Edge

  arXiv (Cornell University) · 2026-04-17

  articleOpen accessSenior author

  Dense wireless deployments face co-channel interference from heterogeneous sources that vary across base stations (gNBs in 5G). While centralized DNN-based approaches to interference mitigation have shown strong performance, deploying and adapting these models across distributed gNBs via federated learning (FL) requires transmitting full model updates each round, resulting in a cost that scales poorly with network density. Parameter-efficient fine-tuning (PEFT) reduces this burden by training and communicating only a small fraction of parameters. While traditionally applied to large foundation models, we adapt Low-Rank Adaptation (LoRA) to temporal convolutional neural network architectures for interference suppression, placing low-rank adapters on the dilated convolutional layers. This placement enables LoRA to learn local interference-specific temporal patterns, while the frozen backbone retains the shared signal extraction capability. These lightweight adapters (5.1\% of backbone parameters) are federated via FedAvg, reducing per-round communication by up to 20$\times$ compared to federating full model updates. We evaluate various PEFT strategies across simulated distributed gNBs with non-IID interference environments. Results show that local LoRA achieves 12.8\% average BER improvement over the frozen backbone, while Fed-LoRA achieves comparable performance (12.6\%). Fed-LoRA outperforms local adaptation on data-starved nodes where federated knowledge transfer compensates for limited samples, all while avoiding the catastrophic degradation observed with full-model FedAvg under heterogeneous conditions.

  PublisherOA PDF

• Who Pays the RENT? Implications of Spatial Inequality for Prediction-Based Allocation Policies

  Proceedings of the AAAI/ACM Conference on AI Ethics and Society · 2025-10-15

  articleOpen accessSenior author

  AI-powered scarce resource allocation policies rely on predictions to target either specific individuals (e.g., high-risk) or settings (e.g., neighborhoods). Recent research on individual-level targeting demonstrates conflicting results; some models show that targeting is not useful when inequality is high, while other work demonstrates potential benefits. To study and reconcile this apparent discrepancy, we develop a stylized framework based on the Mallows model to understand how the spatial distribution of inequality affects the effectiveness of door-to-door outreach policies. We introduce the RENT (Relative Efficiency of Non-Targeting) metric, which we use to assess the effectiveness of targeting approaches compared with neighborhood-based approaches in preventing tenant eviction when high-risk households are more versus less spatially concentrated. We then calibrate the model parameters to eviction court records collected in a medium-sized city in the USA. Results demonstrate considerable gains in the number of high-risk households canvassed through individually targeted policies, even in a highly segregated metro area with concentrated risks of eviction. We conclude that apparent discrepancies in the prior literature can be reconciled by considering 1) the source of deployment costs and 2) the observed versus modeled concentrations of risk. Our results inform the deployment of AI-based solutions in social service provision that account for particular applications and geographies.

  PublisherOA PDFDOI

• Street-Level AI: Are Large Language Models Ready for Real-World Judgments?

  ArXiv.org · 2025-08-11

  preprintOpen accessSenior author

  A surge of recent work explores the ethical and societal implications of large-scale AI models that make "moral" judgments. Much of this literature focuses either on alignment with human judgments through various thought experiments or on the group fairness implications of AI judgments. However, the most immediate and likely use of AI is to help or fully replace the so-called street-level bureaucrats, the individuals deciding to allocate scarce social resources or approve benefits. There is a rich history underlying how principles of local justice determine how society decides on prioritization mechanisms in such domains. In this paper, we examine how well LLM judgments align with human judgments, as well as with socially and politically determined vulnerability scoring systems currently used in the domain of homelessness resource allocation. Crucially, we use real data on those needing services (maintaining strict confidentiality by only using local large models) to perform our analyses. We find that LLM prioritizations are extremely inconsistent in several ways: internally on different runs, between different LLMs, and between LLMs and the vulnerability scoring systems. At the same time, LLMs demonstrate qualitative consistency with lay human judgments in pairwise testing. Findings call into question the readiness of current generation AI systems for naive integration in high-stakes societal decision-making.

  PublisherOA PDFDOI

• Active Geospatial Search for Efficient Tenant Eviction Outreach

  Proceedings of the AAAI Conference on Artificial Intelligence · 2025-04-11

  articleOpen access

  Tenant evictions threaten housing stability and are a major concern for many cities. An open question concerns whether data-driven methods enhance outreach programs that target at-risk tenants to mitigate their risk of eviction. We propose a novel active geospatial search (AGS) modeling framework for this problem. AGS integrates property-level information in a search policy that identifies a sequence of rental units to canvas to both determine their eviction risk and provide support if needed. We propose a hierarchical reinforcement learning approach to learn a search policy for AGS that scales to large urban areas containing thousands of parcels, balancing exploration and exploitation and accounting for travel costs and a budget constraint. Crucially, the search policy adapts online to newly discovered information about evictions. Evaluation using eviction data for a large urban area demonstrates that the proposed framework and algorithmic approach are considerably more effective at sequentially identifying eviction cases than baseline methods.

  PublisherOA PDFDOI

• Reducing Congestion Through Information Design

  World Scientific series in economic theory · 2025-05-01

  book-chapter1st authorCorresponding
  PublisherDOI

• Fixed Points and Stochastic Meritocracies: A Long-Term Perspective

  ArXiv.org · 2025-10-08

  preprintOpen access

  We study group fairness in the context of feedback loops induced by meritocratic selection into programs that themselves confer additional advantage, like college admissions. We introduce a novel stylized inter-generational model for the setting and analyze it in situations where there are no underlying differences between two populations. We show that, when the benefit of the program (or the harm of not getting into it) is completely symmetric, disparities between the two populations will eventually dissipate. However, the time an accumulated advantage takes to dissipate could be significant, and increases substantially as a function of the relative importance of the program in conveying benefits. We also find that significant disparities can arise due to chance even from completely symmetric initial conditions, especially when populations are small. The introduction of even a slight asymmetry, where the group that has accumulated an advantage becomes slightly preferred, leads to a completely different outcome. In these instances, starting from completely symmetric initial conditions, disparities between groups arise stochastically and then persist over time, yielding a permanent advantage for one group. Our analysis precisely characterizes conditions under which disparities persist or diminish, with a particular focus on the role of the scarcity of available spots in the program and its effectiveness. We also present extensive simulations in a richer model that further support our theoretical results in the simpler, stylized model. Our findings are relevant for the design and implementation of algorithmic fairness interventions in similar selection processes.

  PublisherOA PDFDOI

• Who Pays the RENT? Implications of Spatial Inequality for Prediction-Based Allocation Policies

  ArXiv.org · 2025-08-12

  preprintOpen accessSenior author

  AI-powered scarce resource allocation policies rely on predictions to target either specific individuals (e.g., high-risk) or settings (e.g., neighborhoods). Recent research on individual-level targeting demonstrates conflicting results; some models show that targeting is not useful when inequality is high, while other work demonstrates potential benefits. To study and reconcile this apparent discrepancy, we develop a stylized framework based on the Mallows model to understand how the spatial distribution of inequality affects the effectiveness of door-to-door outreach policies. We introduce the RENT (Relative Efficiency of Non-Targeting) metric, which we use to assess the effectiveness of targeting approaches compared with neighborhood-based approaches in preventing tenant eviction when high-risk households are more versus less spatially concentrated. We then calibrate the model parameters to eviction court records collected in a medium-sized city in the USA. Results demonstrate considerable gains in the number of high-risk households canvassed through individually targeted policies, even in a highly segregated metro area with concentrated risks of eviction. We conclude that apparent discrepancies in the prior literature can be reconciled by considering 1) the source of deployment costs and 2) the observed versus modeled concentrations of risk. Our results inform the deployment of AI-based solutions in social service provision that account for particular applications and geographies.

  PublisherOA PDFDOI

• Letter from Chairs

  AI Matters · 2025-10-01

  articleSenior author

  After a short pause, we're excited to announce that AI Matters is returning with a new vision, a broader scope, and an updated format. We envision the newsletter as a useful resource for the three primary communities of ACM SIGAI - AI researchers, professional practitioners, and students - helping everyone stay up to date with developments across the field. AI Matters will continue to advance our mission of promoting the growth and application of AI principles and techniques through computing. We invite you, the members, to take part in this.

  PublisherDOI

• Street-Level AI: Are Large Language Models Ready for Real-World Judgments?

  Proceedings of the AAAI/ACM Conference on AI Ethics and Society · 2025-10-15 · 1 citations

  articleOpen accessSenior author

  A surge of recent work explores the ethical and societal implications of large-scale AI models that make "moral" judgments. Much of this literature focuses either on alignment with human judgments through various thought experiments or on the group fairness implications of AI judgments. However, the most immediate and likely use of AI is to help or fully replace the so-called street-level bureaucrats, the individuals deciding to allocate scarce social resources or approve benefits. There is a rich history underlying how principles of local justice determine how society decides on prioritization mechanisms in such domains. In this paper, we examine how well LLM judgments align with human judgments, as well as with socially and politically determined vulnerability scoring systems currently used in the domain of homelessness resource allocation. Crucially, we use real data on those needing services (maintaining strict confidentiality by only using local large models) to perform our analyses. We find that LLM prioritizations are extremely inconsistent in several ways: internally on different runs, between different LLMs, and between LLMs and the vulnerability scoring systems. At the same time, LLMs demonstrate qualitative consistency with lay human judgments in pairwise testing. Findings call into question the readiness of current generation AI systems for naive integration in high-stakes societal decision-making.

  PublisherOA PDFDOI

Recent grants

• EAGER: AI-DCL: Exploratory research on the use of AI at the intersection of homelessness and child maltreatment

  NSF · $157k · 2021–2022

• CAREER: The Dynamics of Collective Intelligence

  NSF · $228k · 2013–2016

• RI: Small: Modeling Platform Competition: A Multi-Agent Systems Approach

  NSF · $430k · 2015–2019

• RI: Small: Efficient and Just Allocation of Scarce Societal Resources, and Applications to Homelessness

  NSF · $459k · 2019–2021

• EAGER: AI-DCL: Exploratory research on the use of AI at the intersection of homelessness and child maltreatment

  NSF · $300k · 2019–2021

Frequent coauthors

• Yevgeniy Vorobeychik

  17 shared

• Allen Lavoie

  Ministère de l’Environnement, de la Lutte contre les changements climatiques, de la Faune et des Parcs

  16 shared

• Malik Magdon‐Ismail

  Rensselaer Polytechnic Institute

  16 shared

• Patrick J. Fowler

  Washington University in St. Louis

  16 shared

• Andrew Estornell

  15 shared

• Meenal Chhabra

  Research Square (United States)

  11 shared

• Mithun Chakraborty

  Michigan United

  10 shared

• Nicholas Mattei

  Tulane University

  9 shared

Labs

• Institute for Advanced ComputingPI

Education

• Ph.D., Electrical Engineering & Computer Science

  Massachusetts Institute of Technology

  2006

Awards & honors

• AAAI Fellow
• ACM Distinguished Member
• National Science Foundation CAREER Award