paper-with-me

홈 › Papers

Incentives for Collective Innovation

2021-09-04 · Gregorio Curello

Agents exert hidden effort to produce randomly-sized innovations in a technology they share. Returns from using the technology grow as it develops, but so does the opportunity cost of effort, due to an 'exploration-exploitation' trade-off. As monitoring is imperfect, there exists a unique (strongly) symmetric equilibrium, and effort in any equilibrium ceases no later than in the single-agent problem. Small innovations may hurt all agents in the symmetric equilibrium, as they severely reduce effort. Allowing agents to discard innovations increases effort and payoffs, preserving uniqueness. Under natural conditions, payoffs rise above those of all equilibria with forced disclosure.

📄 PDF Abstract BibTeX arXiv:2109.01885

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Learning Incentive Structures for Cooperative Resilience in Multi-Agent Systems under Social Dilemmas

2026-01-29 · Manuela Chacon-Chamorro, Luis Felipe Giraldo, Nicanor Quijano arxiv

Multi-agent social dilemmas, such as the tragedy of the commons, capture settings where individual incentives conflict with collective well-being, making these systems highly vulnerable to collapse under disruptions. In …

Multi-agent Reinforcement Learning

Human diversity fuels collective creativity that large language models cannot simulate or sustain

2026-07-29 · Mengchen Dong, Hiromu Yakura arxiv

Diverse human groups produce diverse ideas, the raw material of innovation. Generative AI challenges this engine twice over: everyday AI assistance may homogenize what diverse people create, and AI-simulated diversity ma…

Causal Foundations of Collective Agency

2026-04-30 · Frederik Hytting Jørgensen, Sebastian Weichwald, Lewis Hammond arxiv

A key challenge for the safety of advanced AI systems is the possibility that multiple simpler agents might inadvertently form a collective agent with capabilities and goals distinct from those of any individual. More ge…

RLHF and IIA: Perverse Incentives

2023-12-02 · Wanqiao Xu, Shi Dong, Xiuyuan Lu, Grace Lam 외

Existing algorithms for reinforcement learning from human feedback (RLHF) can incentivize responses at odds with preferences because they are based on models that assume independence of irrelevant alternatives (IIA). The…

reinforcement-learningReinforcement Learning

Collective Innovation in Groups of Large Language Models

2024-07-07 · Eleni Nisioti, Sebastian Risi, Ida Momennejad, Pierre-Yves Oudeyer 외

Human culture relies on collective innovation: our ability to continuously explore how existing elements in our environment can be combined to create new ones. Language is hypothesized to play a key role in human culture…