![gomez peer zone exploit gomez peer zone exploit](https://i.servimg.com/u/f64/14/64/30/42/tm/2q2rfr10.jpg)
Under these circumstances, the consequences of operating at various positions along the continuum of EETO strategies remain untested.Īn ideal system for investigating EETO strategies in a natural setting is commercial fishing, one of our last remaining hunter–gatherer activities 13. Natural systems are subject to ecological and environmental fluctuations which stochastically modify both the payoff probability of a given option, as well as the portfolio of available options.
![gomez peer zone exploit gomez peer zone exploit](https://fulltimejobfromhome.com/wp-content/uploads/2014/04/gomez-peer-zone-review.jpg)
Although progress has been made in relaxing the assumption of reward stationarity 12, investigation of EETO in the lab remains a profoundly different decision-making setting from the natural systems to which the human EETO-mediating apparatus is adapted 5, 13. Bandit tasks have traditionally assumed that the probability of payoff from a given option (i.e., a machine) is stationary and that the portfolio of options (i.e., the array of machines) from which the subject chooses remains constant 2, 11.
![gomez peer zone exploit gomez peer zone exploit](https://www.pewtrusts.org/-/media/data-visualizations/infographics/2020/03/high-seas-mpa-report/highseas_report_salas_icons_650.jpg)
For example, a subject who eschews exploration could waste time exploiting a machine that only pays off once every 10 trials without ever discovering that a neighboring machine pays off once every three trials. At each choice occasion, the subject decides whether to play the same ‘machine’ or move to another.
#Gomez peer zone exploit series#
However, despite the importance of EETO mediation to adaptive behavior in complex environments 2, 11, many normative and empirical aspects of the problem are poorly understood 2.īehavioral researchers commonly investigate EETO mediation using bandit tasks, where the subject plays a series of one-armed-bandit machines or analogous devices with the goal of devising an EETO strategy with the highest aggregate payoff 5. However, exploration comes at a cost, as it involves increased investment with uncertain outcomes, and time and resources could instead have been invested in exploiting current knowledge to gain immediate, tangible benefits 2, 4, 5 This explore/exploit trade-off (EETO) is pervasive in sequential decision-making settings from financial portfolio blending to machine learning to animal foraging 6, 7, 8, 9, 10, with agents displaying EETO-mediating strategies that place varying emphasis on exploration. Whenever we take a new route to work, for example, we are sampling our environment and adding to a store of information that may increase our long-term benefits (e.g., finding the fastest commute) and/or confer resilience to system dynamics or disturbance should present options become less attractive or unavailable (e.g., knowing alternative routes if traffic is heavy). We conclude that in stochastic natural systems characterized by non-stationary rewards, the role of exploration in buffering against disturbance may be greater than previously thought in humans.Įxploration provides us with information about the surrounding world 1, 2, 3. However, during a major disturbance event which closed the most-utilized fishing grounds, explorers benefited significantly from less-impacted revenues and were also more likely to continue fishing. Leveraging 540,000 vessel position records from 2494 commercial fishing trips along with corresponding revenues, here we find that during undisturbed conditions, there was no relationship between exploration and performance, contrary to theoretical predictions. Although mediation of this trade-off is essential to adaptive behavior and has for decades been assumed to modulate performance, the empirical consequences of human exploratory strategies are unknown beyond laboratory or theoretical settings.
![gomez peer zone exploit gomez peer zone exploit](https://www.mdpi.com/geohazards/geohazards-02-00019/article_deploy/html/images/geohazards-02-00019-g001.png)
Decision-making agents face a fundamental trade-off between exploring new opportunities with risky outcomes versus exploiting familiar options with more certain but potentially suboptimal outcomes.