On Blockchain We Cooperate: An Evolutionary Game Perspective

• URL: http://arxiv.org/abs/2212.05357v3
• Date: Thu, 19 Jan 2023 22:21:26 GMT
• Title: On Blockchain We Cooperate: An Evolutionary Game Perspective
• Authors: Luyao Zhang, Xinyu Tian
• Abstract summary: In this paper, we introduce rationality and game-theoretical solution concepts to study the equilibrium outcomes of consensus protocols. We apply bounded rationality to model agent behavior, and resolve the initial conditions for three different stable equilibria. Our research contributes to the literature across disciplines, including distributed consensus in computer science, game theory in economics on blockchain consensus, evolutionary game theory at the intersection of biology and economics, and cooperative AI with joint insights into computing and social science.
• Score: 0.8566457170664925
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Cooperation is fundamental for human prosperity. Blockchain, as a trust machine, is a cooperative institution in cyberspace that supports cooperation through distributed trust with consensus protocols. While studies in computer science focus on fault tolerance problems with consensus algorithms, economic research utilizes incentive designs to analyze agent behaviors. To achieve cooperation on blockchains, emerging interdisciplinary research introduces rationality and game-theoretical solution concepts to study the equilibrium outcomes of various consensus protocols. However, existing studies do not consider the possibility for agents to learn from historical observations. Therefore, we abstract a general consensus protocol as a dynamic game environment, apply a solution concept of bounded rationality to model agent behavior, and resolve the initial conditions for three different stable equilibria. In our game, agents imitatively learn the global history in an evolutionary process toward equilibria, for which we evaluate the outcomes from both computing and economic perspectives in terms of safety, liveness, validity, and social welfare. Our research contributes to the literature across disciplines, including distributed consensus in computer science, game theory in economics on blockchain consensus, evolutionary game theory at the intersection of biology and economics, bounded rationality at the interplay between psychology and economics, and cooperative AI with joint insights into computing and social science. Finally, we discuss that future protocol design can better achieve the most desired outcomes of our honest stable equilibria by increasing the reward-punishment ratio and lowering both the cost-punishment ratio and the pivotality rate.

Related papers

• The State-Action-Reward-State-Action Algorithm in Spatial Prisoner's Dilemma Game [0.0]
  Reinforcement learning provides a suitable framework for studying evolutionary game theory. We employ the State-Action-Reward-State-Action algorithm as the decision-making mechanism for individuals in evolutionary game theory. We evaluate the impact of SARSA on cooperation rates by analyzing variations in rewards and the distribution of cooperators and defectors within the network.
  arXiv  Detail & Related papers  (2024-06-25T07:21:35Z)
• Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents [101.17919953243107]
  GovSim is a generative simulation platform designed to study strategic interactions and cooperative decision-making in large language models (LLMs) We find that all but the most powerful LLM agents fail to achieve a sustainable equilibrium in GovSim, with the highest survival rate below 54%. We show that agents that leverage "Universalization"-based reasoning, a theory of moral thinking, are able to achieve significantly better sustainability.
  arXiv  Detail & Related papers  (2024-04-25T15:59:16Z)
• Multi-Agent Dynamic Relational Reasoning for Social Robot Navigation [55.65482030032804]
  Social robot navigation can be helpful in various contexts of daily life but requires safe human-robot interactions and efficient trajectory planning. We propose a systematic relational reasoning approach with explicit inference of the underlying dynamically evolving relational structures. Our approach infers dynamically evolving relation graphs and hypergraphs to capture the evolution of relations, which the trajectory predictor employs to generate future states.
  arXiv  Detail & Related papers  (2024-01-22T18:58:22Z)
• Social Motion Prediction with Cognitive Hierarchies [19.71780279070757]
  We introduce a new benchmark, a novel formulation, and a cognition-inspired framework. We present Wusi, a 3D multi-person motion dataset under the context of team sports. We develop a cognitive hierarchy framework to predict strategic human social interactions.
  arXiv  Detail & Related papers  (2023-11-08T14:51:17Z)
• Tackling Cooperative Incompatibility for Zero-Shot Human-AI Coordination [36.33334853998621]
  We introduce the Cooperative Open-ended LEarning (COLE) framework to solve cooperative incompatibility in learning. COLE formulates open-ended objectives in cooperative games with two players using perspectives of graph theory to evaluate and pinpoint the cooperative capacity of each strategy. We show that COLE could effectively overcome the cooperative incompatibility from theoretical and empirical analysis.
  arXiv  Detail & Related papers  (2023-06-05T16:51:38Z)
• The Design Principle of Blockchain: An Initiative for the SoK of SoKs [0.0]
  Decentralized AI can empower to be more trustworthy by creating a decentralized trust of privacy, security, and audibility. However, systematic studies on the design principle of AI as a trust engine for an integrated society of Cyber-Physical-System are still absent.
  arXiv  Detail & Related papers  (2023-01-01T21:57:25Z)
• Finding General Equilibria in Many-Agent Economic Simulations Using Deep Reinforcement Learning [72.23843557783533]
  We show that deep reinforcement learning can discover stable solutions that are epsilon-Nash equilibria for a meta-game over agent types. Our approach is more flexible and does not need unrealistic assumptions, e.g., market clearing. We demonstrate our approach in real-business-cycle models, a representative family of DGE models, with 100 worker-consumers, 10 firms, and a government who taxes and redistributes.
  arXiv  Detail & Related papers  (2022-01-03T17:00:17Z)
• The AI Economist: Optimal Economic Policy Design via Two-level Deep Reinforcement Learning [126.37520136341094]
  We show that machine-learning-based economic simulation is a powerful policy and mechanism design framework. The AI Economist is a two-level, deep RL framework that trains both agents and a social planner who co-adapt. In simple one-step economies, the AI Economist recovers the optimal tax policy of economic theory.
  arXiv  Detail & Related papers  (2021-08-05T17:42:35Z)
• Cooperation and Reputation Dynamics with Reinforcement Learning [6.219565750197311]
  We show how reputations can be used as a way to establish trust and cooperation. We propose two mechanisms to alleviate convergence to undesirable equilibria. We show how our results relate to the literature in Evolutionary Game Theory.
  arXiv  Detail & Related papers  (2021-02-15T12:48:56Z)
• On Information Asymmetry in Competitive Multi-Agent Reinforcement Learning: Convergence and Optimality [78.76529463321374]
  We study the system of interacting non-cooperative two Q-learning agents. We show that this information asymmetry can lead to a stable outcome of population learning.
  arXiv  Detail & Related papers  (2020-10-21T11:19:53Z)
• Decentralized Reinforcement Learning: Global Decision-Making via Local Economic Transactions [80.49176924360499]
  We establish a framework for directing a society of simple, specialized, self-interested agents to solve sequential decision problems. We derive a class of decentralized reinforcement learning algorithms. We demonstrate the potential advantages of a society's inherent modular structure for more efficient transfer learning.
  arXiv  Detail & Related papers  (2020-07-05T16:41:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.