Cycles of cooperation and defection in imperfect learning

  title={Cycles of cooperation and defection in imperfect learning},
  author={Tobias Galla},
  journal={Journal of Statistical Mechanics: Theory and Experiment},
  • T. Galla
  • Published 23 January 2011
  • Materials Science
  • Journal of Statistical Mechanics: Theory and Experiment
Commentary on `Cycles of cooperation and defection in imperfect learning', by Tobias Galla, 2011 J. Stat. Mech. P08007. 

Intrinsic Fluctuations of Reinforcement Learning Promote Cooperation

This work considers the widely used temporal-difference reinforcement learning algorithm with epsilon-greedy exploration in the classic environment of an iterated Prisoner's dilemma with one-period memory and demonstrates which and how individual elements of the multi-agent learning setting lead to cooperation.

Learning dynamics in public goods games.

It is shown that coherent cycles may emerge driven by noise in the adaptation dynamics of multiplayer public goods games, not too dissimilar from cyclic strategy switching observed in experiments of behavioral game theory.

Limit Cycles Sparked by Mutation in the Repeated Prisoner's Dilemma

Stable oscillations are a robust aspect of a world of ALLC, ALLD, and costly TFT; the existence of cycles does not depend on the details of assumptions of how mutation is implemented.

Fence-sitters Protect Cooperation in Complex Networks

A vectorial formulation to derive three classes of individuals' payoff analytically, which indicates that the fence-sitters' role is nontrivial in the complex topologies, which protects cooperation in an indirect way.

Effects of noise on convergent game-learning dynamics

Stochastic effects on the lagging anchor dynamics, a reinforcement learning algorithm used to learn successful strategies in iterated games, are studied and it is found that the system can exhibit quasicycles, driven by intrinsic noise.

Evolution of global contribution in multi-level threshold public goods games with insurance compensation

Understanding voluntary contribution in threshold public goods games has important practical implications. To improve contributions and provision frequency, free-rider problem and assurance problem

Towards a Taxonomy of Learning Dynamics in 2 × 2 Games

Learning would be a convincing method to achieve coordination on an equilibrium. But does learning converge, and to what? We answer this question in generic 2-player, 2-strategy games, using

Intrinsically motivated reinforcement learning in socio-economic systems: The dynamical analysis

  • A. ZgonnikovI. Lubashevsky
  • Economics
    2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)
  • 2013
It is argued that the effects of human intrinsic motivation in particular and bounded rationality in general may appear dominant in complex socio-economic systems and therefore deserve much attention in the formal models of such systems.

Coordination problems on networks revisited: statics and dynamics

  • L. Dall’Asta
  • Economics
    Journal of Statistical Mechanics: Theory and Experiment
  • 2021
It is shown that well beyond the instability region, full coordination is still globally stochastically stable, however equilibrium selection processes with low stochasticity or strong memory effects can be prevented from achieving full coordination by being trapped into a large set of locally stable Nash equilibria at low/medium coordination.

Unstable Dynamics of Adaptation in unknown Environment due to Novelty seeking

The results give evidence to the hypothesis that the intrinsic motives of agents should be paid no less attention than the extrinsic ones in the models of complex socio-economic systems.



How Did Cooperative Behavior Evolve?

Scientists are applying evolutionary game theory to quantify cooperation and predict behavioral outcomes under different circumstances, and hope to gain a clearer sense of the rules that govern complex societies.

Evolutionary cycles of cooperation and defection.

This work analyzes the evolutionary dynamics of three simple strategies for the repeated prisoner's dilemma: always defect (ALLD), always cooperate (ALLC), and tit-for-tat (TFT) and observes evolutionary oscillations among all three strategies.

Behavioral Game Theory: Experiments in Strategic Interaction

The book describes experiments in Strategic Interaction using game theory as a guide to solving social problems.

Algorithmic game theory

A new era of theoretical computer science addresses fundamental problems about auctions, networks, and human behavior in a bid to solve the challenges of 21st Century finance.

Volunteering leads to rock–paper–scissors dynamics in a public goods game

It is shown experimentally that volunteering generates dynamics in public goods games and that manipulating initial conditions can produce each predicted direction, and that cooperation is perpetuated at a substantial level.

Evolutionary dynamics, intrinsic noise, and cycles of cooperation.

Analysis of the stochastic evolutionary dynamics of finite populations of players interacting in a repeated prisoner's dilemma game shows that a mechanism of amplification of demographic noise can give rise to coherent oscillations in parameter regimes where deterministic descriptions converge to fixed points with complex eigenvalues.

Effect of memory on the prisoner's dilemma game in a square lattice.

It is interesting to note that memory makes cooperators immune from temptation, and the strength of protection reaches its maximal value only for a moderate memory effect.

The Theory of Learning in Games

The evolution of cooperation.

A model is developed based on the concept of an evolutionarily stable strategy in the context of the Prisoner's Dilemma game to show how cooperation based on reciprocity can get started in an asocial world, can thrive while interacting with a wide range of other strategies, and can resist invasion once fully established.