Moran
Koren

Summary: This paper analyzes how pre-screening stages ("gatekeepers") in hiring can paradoxically lower the average quality of the applicant pool by altering candidates' self-selection incentives and perpetuating historical biases.

Tools & Methodologies: Game Theory, Algorithmic Screening, Information Design, Self-Selection Models.

Summary: We develop a practical, strategy-proof mechanism for resolving ownership deadlocks in contexts like joint ventures or supply chain partnerships, ensuring efficient and fair outcomes.

Tools & Methodologies: Mechanism Design, Game Theory, Corporate Finance.

Summary: Examines Israel's food system through the lens of the Mediterranean Diet, combining FAO Food Balance Sheets and CBS data to quantify import-dependency ratios and alignment with MD guidelines. Documents heavy import dependency in cereals, fish, nuts, and added fats, and shows that the current supply deviates from MD recommendations — notably undersupplying plant-based proteins. Argues for a shift toward locally produced, plant-based food and feed as a joint lever for public health, food security, resilience, and sustainability.

Tools & Methodologies: Food Systems Analysis, Operations Management, Public Policy, Resilience, Mediterranean Diet, Import Dependency.

Summary: This work examines how information from individual consumers is aggregated in settings like group buying or crowdfunding, analyzing the efficiency of collective decision-making under uncertainty.

Tools & Methodologies: Information Aggregation, Game Theory, Collective Action.

Summary: This research connects contract theory with machine learning by introducing algorithms that enable a principal to learn the optimal contract to offer an agent whose cost structure is initially unknown.

Tools & Methodologies: Algorithmic Mechanism Design, Contract Theory, Machine Learning, Online Learning.

Summary: This paper investigates how a manager can use dynamic pricing as an operational lever to learn consumer demand, identifying conditions under which a firm can overcome informational herds and guarantee long-run learning.

Tools & Methodologies: Social Learning, Game Theory, Dynamic Pricing, Information Design.

Summary: We demonstrate a critical managerial tradeoff by showing that operational policies designed to be "informationally efficient" can paradoxically reduce short-term profitability and overall welfare.

Tools & Methodologies: Information Economics, Game Theory, Market Efficiency.

Summary: Proposes a verification-first protocol for doing economic theory with large language models, instantiated as three reusable methods that differ only in how the work is checked: a single disciplined pass, an adversarial prover–verifier pair, and a structured multi-agent project with a reviewer gate. Demonstrated on designing a Groves/Pigouvian incentive mechanism for the Gans–Kominers eigengrade model of grade inflation. The methodological takeaway: external verification, not model capability, is the design variable.

Tools & Methodologies: LLM-Assisted Theory, Mechanism Design, Adversarial Verification, Multi-Agent Workflows.

Summary: A model of competition in common value markets whose equilibrium predictions are tested in a large-scale online experiment, measuring how competitive pressure trades off decision speed against the quality of information aggregation. Based on Yuval Peled's MSc thesis.

Tools & Methodologies: Game Theory, Online Experiments, Information Aggregation, Common-Value Markets.

Summary: Tests whether LLM agents do what they say in Settlers of Catan, a multi-player negotiation game: how consistent are stated intentions with actual play, and which grounding mechanisms close the gap.

Tools & Methodologies: LLM Agents, Multi-Agent Systems, Game Theory, Agent Evaluation.

Summary: Defines Goldstein (stationary-point) equilibria for two-player games with Lipschitz payoffs and gives procedures to certify them.

Tools & Methodologies: Game Theory, Nonsmooth Optimization, Equilibrium Computation, Certification.

Summary: Revisits fair solutions to repetitive single-machine scheduling: n clients each submit one job per day over q days, and the schedules must give every client a quality of service (e.g., total waiting or completion time over the horizon) above a threshold. Resolves several open questions on the problem's computational complexity and on the price of fairness, and maps directions for further work.

Tools & Methodologies: Scheduling Theory, Computational Complexity, Fairness, Price of Fairness, Approximation Algorithms.

Summary: Develops a fairness-constrained actor–critic method for multi-agent reinforcement learning; Koren leads the economic-theory aspects of the work.

Tools & Methodologies: Multi-Agent Reinforcement Learning, Fairness, Actor–Critic Methods.

Summary: A sole-authored pedagogical approach that teaches game theory by having students design and build strategic agents, turning abstract equilibrium concepts into hands-on engineering of the players themselves.

Tools & Methodologies: Game Theory, Pedagogy, Agent Design.

Summary: This work designs mechanisms that balance information acquisition and strategic incentives in dynamic allocation markets (e.g., organ/platform markets). It proposes a novel batch-based voting mechanism to mitigate herd behavior and improve allocation efficiency.

Tools & Methodologies: Mechanism Design, Game Theory, Social Learning, Market Design, Simulations.

Summary: Characterizes when sequential fundraising with risk-sharing generates inefficiency and provides implementable design fixes; delivers testable comparative statics.

Tools & Methodologies: Game Theory, Mechanism Design, Crowdfunding, Social Learning.

Summary: Models the proposed switch to batch offering in organ allocation as a common-value variant of the Dynkin (1969) stopping game. Agents compete over an offering of unknown value and receive private signals over time; each must trade off waiting for more information against losing the offer to a competitor. We prove existence of a symmetric threshold equilibrium in the binary-signal case, characterize the asymptotic barrier strategy (its velocity rises with signal accuracy), and show numerically that larger batches raise the probability of a mistake, lower the probability of waste, and shorten time-to-resolution in the good state. Translating the parameters into policy levers (batch size, diagnostic quality, organ preservation, supply), we find that signal accuracy dominates batch size for welfare — the success of batch offering depends less on how many centers compete and more on how good their information is.

Tools & Methodologies: Stopping Games, Common-Value Auctions, Mechanism Design, Organ Allocation, Numerical Equilibrium Analysis.

Summary: Introduces a graph-theoretic resettlement mechanism that partitions a weighted social network into capacity-constrained groups while minimizing severed social ties; proves the cut-minimizing assignment is stable and develops a scalable cycle-based heuristic with strong empirical performance.

Tools & Methodologies: Network Optimization, Matching & Allocation, Minimum-Cut Partitioning, Refugee/IDP Resettlement, Heuristics.

Summary: Models decentralized assignment of agents to localities over a weighted social network as an exact potential game with threshold-linear congestion; establishes existence and convergence of pure Nash equilibria and tight price-of-anarchy bounds for socially-aware resettlement. Extends S. Horowitz's BSc final project.

Tools & Methodologies: Potential Games, Congestion Games, Price of Anarchy, Decentralized Mechanism Design.

Summary: Per-offer kidney acceptance fell by roughly half after KAS250 replaced DSA/regional allocation with a 250-nautical-mile circle in March 2021. Using 67 million OPTN STAR offer decisions (2018–19 vs. April 2021–April 2023), the paper decomposes the decline into a mechanical match-run-depth component and genuine center reluctance. At rank 1, where depth cannot operate, acceptance fell from 24.1% to 14.1% (adjusted OR 0.53), and provisional-yes-to-final conversion fell 42%; genuine reluctance (~40%) exceeds what a naive depth adjustment implies (~19%). Broader sharing independently raised discard, concentrated in DCD kidneys.

Tools & Methodologies: Registry Data (OPTN STAR), Before/After and Difference-in-Differences Analysis, Logistic Regression, Organ Allocation Policy.

Summary: This student-led publication serves as a proof-of-concept for translating a core research agenda into a productive, student-driven research program. It directly validates theoretical predictions from earlier work.

Tools & Methodologies: Online Experiments, Behavioral Economics, Game Theory, Crowdfunding Platforms.

Summary: Develops entry-aware classification under endogenous applicant pools and proposes screening rules that are robust to self-selection.

Tools & Methodologies: Machine Learning, Algorithmic Game Theory, Causal Inference, Fairness in AI.