60 papers
This paper proposes a tractable two-layer framework combining a Hidden Markov Model for inferring rival energy states and a Deep Q-Network for decision-making to optimize 2026 Formula 1 energy strategies under partial observability, specifically addressing the "counter-harvest trap" where opponents deliberately mask their deployment signals.
This paper proposes a principled feedback control method combining bucketized hysteresis and proportional feedback to achieve stable, accurate, and scalable budget pacing for online advertising campaigns, particularly improving performance for small-budget scenarios by significantly reducing pacing errors and volatility compared to existing baselines.
This paper models benchmarking as a strategic game to demonstrate that current evaluation protocols fail to establish a stable equilibrium, whereas a proposed "tune-before-test" protocol successfully incentivizes developers to improve latent model quality by inducing a unique Nash equilibrium.
This paper proposes a lightweight, fully online Model Predictive Control framework for brand auction ads that utilizes isotonic regression to build monotonic bid models from streaming data, thereby improving spend efficiency and cost control without the need for complex machine learning models.
This paper introduces Deep Incentive Design (DID), a novel framework that employs game-agnostic differentiable equilibrium blocks (DEBs) to enable the automated, neural-network-based design of multi-agent interactions across diverse economic and computer science tasks, effectively solving complex incentive problems by handling a wide range of game scales and contexts.
This paper demonstrates that optimizing for objectives like profit or crime reduction while maintaining a zero Average Treatment Effect (ATE) can mask significant unfairness driven by confounding, thereby arguing that fairness regulations must shift from evaluating aggregate decision-level outcomes to scrutinizing model-level causal mechanisms.
This paper establishes the exact minimax regret rate for contextual bilateral trade under heavy-tailed valuations with infinite variance by extending self-bounding properties to real-valued settings and combining them with truncated-mean estimation to derive an epoch-based algorithm that interpolates between classical nonparametric and trivial linear rates.
This paper employs an AI-guided evolutionary search framework to identify a new worst-case distribution that establishes a lower bound of 2.0749 for the approximation ratio of the Random-Offerer mechanism in bilateral trade, surpassing previous conjectures and known counterexamples.
This paper introduces NePPO, a novel multi-agent reinforcement learning pipeline that computes approximate Nash equilibria in general-sum games by learning a player-independent potential function to transform the mixed cooperative-competitive environment into an approximating cooperative game.
This paper proposes a novel level-2 inverse game framework that infers agents' estimates of each other's objectives to address the limitations of traditional level-1 methods in decentralized scenarios, demonstrating through theory and experiments that accounting for these mutual misalignments is crucial for accurately predicting strategic interactions.
This paper proves that determining whether a computable strategy guarantees a win in the Yu-Gi-Oh! Trading Card Game is an undecidable problem that is actually -complete, demonstrating this by constructing legal decks capable of reducing the Halting Problem and the set of countable well orders to the game's winning condition.
This paper investigates Instant Runoff Voting (IRV) on unweighted graphs by characterizing exclusion zones and distortion, demonstrating that while testing and finding minimum exclusion zones are computationally hard in general graphs, they become polynomial-time solvable on trees through a novel dynamic programming approach, while also establishing hardness for broader classes of rank-based elimination rules.
This paper introduces Code-Space Response Oracles (CSRO), a novel framework that replaces black-box deep reinforcement learning oracles with Large Language Models to generate human-readable, interpretable multi-agent policies as code, achieving competitive performance while enabling the discovery of complex, explainable strategies.
This paper proposes a machine learning-enhanced Combinatorial Clock Exchange mechanism for local energy markets that simplifies preference elicitation for prosumers by allowing them to report preferred product packages at posted prices, thereby overcoming cognitive limitations and achieving efficient market convergence in approximately 15 iterations.
This paper introduces a game-theoretic evaluation framework using quantal response equilibrium to measure the strategic sophistication of large language models by deriving closed-form equilibria, estimating continuous rationality parameters calibrated against human data, and validating the approach across thousands of games to reveal both model capabilities and significant sensitivities to prompt framing.
This paper extends Causal Normal Form Games to sequential settings by introducing Sequential Causal Multi-Agent Systems, but its comprehensive theoretical and empirical analysis reveals that, under standard rational assumptions, these causal frameworks offer no welfare advantage over classical Stackelberg equilibrium, thereby highlighting a fundamental incompatibility between rational choice and causal reasoning benefits in current game-theoretic models.
This paper resolves a key open question from Gorantla et al. by deriving explicit upper bounds on the number of item copies required to guarantee envy-free allocations for arbitrary numbers of agent groups and item types, using a novel constructive technique that extends to both chores and continuous fair division settings.
This paper proposes a "Test-then-Punish" framework that sustains cooperation in discounted infinitely repeated games with imperfect monitoring by embedding statistical hypothesis testing into strategic behavior, allowing players to detect deviations and enforce a Folk theorem-type result through either anytime valid sequential tests or batch-based testing.
This paper extends the temporal network creation game model by allowing agents to flexibly assign time labels to purchased edges rather than using predetermined ones, analyzing the resulting existence of Nash equilibria and bounds on the Price of Anarchy and Stability under various reachability and cost scenarios.
This paper presents the first polynomial-time algorithms that guarantee allocations of indivisible goods are simultaneously envy-free up to one good (EF1) and fractionally Pareto optimal (fPO) under balanced constraints, specifically for agents with personalized bivalued valuations or at most two distinct valuation types.