This article documents a meta-analysis of

This paper seeks to evaluate the impact of the removal of restrictions (partial and complete) imposed during COVID-19-induced lockdowns on property offences such as robbery, burglary, and theft during the milder wave one and the more severe wave two of the pandemic in 2020 and 2021, respectively. Using 10-year data of the daily counts of crimes, the authors adopt an auto-regressive neural networks method to make counterfactual predictions of crimes, representing a scenario without the pandemic-induced lockdowns. The difference between the actual and forecast is the causal impact of the lockdown in all phases. Further, the research uses Google Mobility Community Reports to measure mobility. The analysis has been done at two levels: first, for the state of Tamil Nadu, which has a sizeable rural landscape, and second for Chennai, the largest metropolitan city with an urban populace. During the pandemic-induced lockdown in wave one, there was a steep decline in the incidence of property offences. On removing restrictions, the cases soared above the counterfactual predicted counts. In wave two, despite the higher severity and fatality in the COVID-19 pandemic, a similar trend of fall and rise in property cases was observed. However, the drop in mobility was less substantial, and the increase in the magnitude of property offences was more significant in wave two than in wave one. The overall trend of fluctuations is related to mobility during various phases of restrictions in the pandemic. When most curbs were removed, there was a surge in robberies in Tamil Nadu and Chennai after adjusting for mobility. This trend highlights the effective increase in crime due to pandemic-related economic and social consequences. Further, the research enables law enforcement to strengthen preventive crime work in similar situations, when most curbs are removed after a pandemic or other unanticipated scenarios.

No abstract

In this paper we propose the Augmented-UCB (AugUCB) algorithm for a fixed-budget version of the thresholding bandit problem (TBP), where the objective is to identify a set of arms whose quality is above a threshold. A key feature of AugUCB is that it uses both mean and variance estimates to eliminate arms that have been sufficiently explored; to the best of our knowledge this is the first algorithm to employ such an approach for the considered TBP. Theoretically, we obtain an upper bound on the loss (probability of mis-classification) incurred by AugUCB. Although UCBEV in literature provides a better guarantee, it is important to emphasize that UCBEV has access to problem complexity (whose computation requires arms' mean and variances), and hence is not realistic in practice; this is in contrast to AugUCB whose implementation does not require any such complexity inputs. We conduct extensive simulation experiments to validate the performance of AugUCB. Through our simulation work, we establish that AugUCB, owing to its utilization of variance estimates, performs significantly better than the state-of-the-art APT, CSAR and other non variance-based algorithms.