Learning Through Reinforcement and Replicator Dynamics

By Tilman Börgers, Rajiv Sarin. 1997

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Lenient Multi-Agent Deep Reinforcement Learning (2017, background): “However, leniency has been shown to increase the likelihood of convergence towards the globally optimal solution in stateless cooperation games for reinforcement learning agents [2, 21, 16] which should not come as a surprise given that both reinforcement learning and evolutionary algorithms converg”
Evolutionary Dynamics of Multi-Agent Learning: A Survey (2015, methodology): “The link between these two fields is not only an intuitive one, but was made formal with the proof that the continuous time limit of Cross learning converges to the replicator dynamics (Börgers & Sarin, 1997).”
Individual and cultural learning in stag hunt games with multiple actions (2010, background): “If for all j ∈ E , inequality (7) holds and xj < Tj, then the self interested action S has a higher payoff than any of the effective actions.”
Online learning control by association and reinforcement (2000, background): “Dynamic programming has been applied in different fields of engineering, operations research, economics, and so on for many years [2], [5], [6], [22].”
Generalised weakened fictitious play (2006, result): “These results, as well as those of Börgers and Sarin (1997) on a different version of reinforcement learning, are obtained by exploiting a relationship with the replicator dynamics of evolutionary game theory (Taylor and Jonker, 1978), analogously to the relationship between generalised weakened…”
Autocurricula and the Emergence of Innovation from Social Interaction: A Manifesto for Multi-Agent Intelligence Research (2019, background): “Many of the principles remain the same regardless of whether evolution or learning dominate (Bloembergen et al., 2015; Börgers and Sarin, 1997; Such et al., 2017).”
An Economist's Perspective on Probability Matching (2000, background): “Second, to familiarise the reader with the different theoretical approaches to behaviour in such circumstances, and to focus on the differences in predictions between these theories in light of recent advances (such as Borgers & Sarin (1993, 1995) , and Borgers, Morales and Sarin (1997)).”
Analyzing Complex Strategic Interactions in Multi-Agent Systems (2002, background): “Alternative models in which learning at the individual level leads to replicator dynamics has been discussed in ( Borgers & Sarin 1997 ).”
Neural Replicator Dynamics (2019, background): “5] and max-logit-gap β uniformly in the interval [5 , 50] .”
Ideas are not replicators but minds are (2004, background): “Evolutionary theory has been also applied in general terms to units of culture referred to as memes (Aunger 2000; Blackmore 1999, 2000; Dawkins 1976; Dennett 1995; Gabora 1996) as well as to the analysis of growth and change in economics ( Borger and Sarin 1997; Borkar et al. 1998; Metcalfe 2001; R”
Deterministic Approximation of Stochastic Evolution in Games (2003, background): “If the threshold distributions are uniform, with a support [ai(x); bi(x)] that covers the range of possible payo® di®erences between any two pure strategies in player role i, then it is easily shown that the mean vector- ̄eld is twice that in equation (11), and hence one obtains the Taylor (1979) an”
Naive Reinforcement Learning With Endogenous Aspirations (2000, background): “Another reason for the recent interest in reinforcement learning among economists is that there is a close analogy between reinforcement learning and dynamic processes studied in evolutionary game theory (see Börgers and Sarin, 1997).”
A Survey of Learning in Multiagent Environments: Dealing with Non-Stationarity (2017, background): “Initial work bringing this field towards multi-agent learning algorithms has established the formal link between the simple reinforcement learning algorithm cross learning and the replicator dynamics, a central concept in evolutionary game theory (Börgers and Sarin, 1997).”