Pith. sign in

Learning Optimal Contracts: How to Exploit Small Action Spaces

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

We study principal-agent problems in which a principal commits to an outcome-dependent payment scheme -- called contract -- in order to induce an agent to take a costly, unobservable action leading to favorable outcomes. We consider a generalization of the classical (single-round) version of the problem in which the principal interacts with the agent by committing to contracts over multiple rounds. The principal has no information about the agent, and they have to learn an optimal contract by only observing the outcome realized at each round. We focus on settings in which the size of the agent's action space is small. We design an algorithm that learns an approximately-optimal contract with high probability in a number of rounds polynomial in the size of the outcome space, when the number of actions is constant. Our algorithm solves an open problem by Zhu et al.[2022]. Moreover, it can also be employed to provide a $\tilde{\mathcal{O}}(T^{4/5})$ regret bound in the related online learning setting in which the principal aims at maximizing their cumulative utility, thus considerably improving previously-known regret bounds.

fields

cs.GT 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

Algorithmic Fair Contracts

cs.GT · 2025-07-15 · conditional · novelty 8.0

Envy-free task-contract allocations always exist, yet maximizing principal revenue under envy-freeness is NP-hard to approximate in general; polynomial algorithms exist for constant tasks, and FPTAS for epsilon-EF and EF1 with constant agents.

citing papers explorer

Showing 1 of 1 citing paper.

  • Algorithmic Fair Contracts cs.GT · 2025-07-15 · conditional · none · ref 1 · internal anchor

    Envy-free task-contract allocations always exist, yet maximizing principal revenue under envy-freeness is NP-hard to approximate in general; polynomial algorithms exist for constant tasks, and FPTAS for epsilon-EF and EF1 with constant agents.