Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userOn Statistical Discrimination as a Failure of Social Learning: A Multi-Armed Bandit Approach
197
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
We analyze statistical discrimination in hiring markets using a multi-armed bandit model. Myopic firms face workers arriving with heterogeneous observable characteristics. The association between the workers skill and characteristics is unknown ex ante; thus, firms need to learn it. Laissez-faire causes perpetual underestimation: minority workers are rarely hired, and therefore, underestimation towards them tends to persist. Even a slight population-ratio imbalance frequently produces perpetual underestimation. We propose two policy solutions: a novel subsidy rule (the hybrid mechanism) and the Rooney Rule. Our results indicate that temporary affirmative actions effectively mitigate discrimination caused by insufficient data.
rate research
Read More
We consider a multi-round auction setting motivated by pay-per-click auctions for Internet advertising. In each round the auctioneer selects an advertiser and shows her ad, which is then either clicked or not. An advertiser derives value from clicks; the value of a click is her private information. Initially, neither the auctioneer nor the advertisers have any information about the likelihood of clicks on the advertisements. The auctioneers goal is to design a (dominant strategies) truthful mechanism that (approximately) maximizes the social welfare. If the advertisers bid their true private values, our problem is equivalent to the multi-armed bandit problem, and thus can be viewed as a strategic version of the latter. In particular, for both problems the quality of an algorithm can be characterized by regret, the difference in social welfare between the algorithm and the benchmark which always selects the same best advertisement. We investigate how the design of multi-armed bandit algorithms is affected by the restriction that the resulting mechanism must be truthful. We find that truthful mechanisms have certain strong structural properties -- essentially, they must separate exploration from exploitation -- and they incur much higher regret than the optimal multi-armed bandit algorithms. Moreover, we provide a truthful mechanism which (essentially) matches our lower bound on regret.
We study the implications of endogenous pricing for learning and welfare in the classic herding model . When prices are determined exogenously, it is known that learning occurs if and only if signals are unbounded. By contrast, we show that learning can occur when signals are bounded as long as non-conformism among consumers is scarce. More formally, learning happens if and only if signals exhibit the vanishing likelihood property introduced bellow. We discuss the implications of our results for potential market failure in the context of Schumpeterian growth with uncertainty over the value of innovations.
Base station (BS) placement in mobile networks is critical to the efficient use of resources in any communication system and one of the main factors that determines the quality of communication. Although there is ample literature on the optimum placement of BSs for sub-6 GHz bands, channel propagation characteristics, such as penetration loss, are notably different in millimeter-wave (mmWave) bands than in sub-6 GHz bands. Therefore, designated solutions are needed for mmWave systems to have reliable quality of service (QoS) assessment. This article proposes a multi-armed bandit (MAB) learning approach for the mmWave BS placement problem. The proposed solution performs viewshed analysis to identify the areas that are visible to a given BS location by considering the 3D geometry of the outdoor environments. Coverage probability, which is used as the QoS metric, is calculated using the appropriate path loss model depending on the viewshed analysis and a probabilistic blockage model and then fed to the MAB learning mechanism. The optimum BS location is then determined based on the expected reward that the candidate locations attain at the end of the training process. Unlike the optimization-based techniques, this method can capture the time-varying behavior of the channel and find the optimal BS locations that maximize long-term performance.
The successful launch of electric vehicles (EVs) depends critically on the availability of convenient and economic charging facilities. The problem of scheduling of large-scale charging of EVs by a service provider is considered. A Markov decision process model is introduced in which EVs arrive randomly at a charging facility with random demand and completion deadlines. The service provider faces random charging costs, convex non-completion penalties, and a peak power constraint that limits the maximum number of simultaneous activation of EV chargers. Formulated as a restless multi-armed bandit problem, the EV charging problem is shown to be indexable. A closed-form expression of the Whittles index is obtained for the case when the charging costs are constant. The Whittles index policy, however, is not optimal in general. An enhancement of the Whittles index policy based on spatial interchange according to the less laxity and longer processing time principle is presented. The proposed policy outperforms existing charging algorithms, especially when the charging costs are time varying.
332 -
Vineeth S. Varma
, Samson Lasaulce
, Julien Mounthanyvong andn Irinel-Constantin Morarescu
2020
In this paper, we consider a network of consumers who are under the combined influence of their neighbors and external influencing entities (the marketers). The consumers opinion follows a hybrid dynamics whose opinion jumps are due to the marketing campaigns. By using the relevant static game model proposed recently in [1], we prove that although the marketers are in competition and therefore create tension in the network, the network reaches a consensus. Exploiting this key result, we propose a coopetition marketing strategy which combines the one-shot Nash equilibrium actions and a policy of no advertising. Under reasonable sufficient conditions, it is proved that the proposed coopetition strategy profile Pareto-dominates the one-shot Nash equilibrium strategy. This is a very encouraging result to tackle the much more challenging problem of designing Pareto-optimal and equilibrium strategies for the considered dynamical marketing game.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
