Search for: All records

We fully characterize the nonasymptotic minimax separation rate for sparse signal detection in the Gaussian sequence model with $p$ equicorrelated observations, generalizing a result of Collier, Comminges and Tsybakov. As a consequence of the rate characterization, we find that strong correlation is a blessing, moderate correlation is a curse and weak correlation is irrelevant. Moreover, the threshold correlation level yielding a blessing exhibits phase transitions at the $\sqrt{p}$ and $p-\sqrt{p}$ sparsity levels. We also establish the emergence of new phase transitions in the minimax separation rate with a subtle dependence on the correlation level. Additionally, we study group structured correlations and derive the minimax separation rate in a model including multiple random effects. The group structure turns out to fundamentally change the detection problem from the equicorrelated case and different phenomena appear in the separation rate.

 

Data on population movements can be helpful in designing targeted policy responses to curb epidemic spread. However, it is not clear how to exactly leverage such data and how valuable they might be for the control of epidemics. To explore these questions, we study a spatial epidemic model that explicitly accounts for population movements and propose an optimization framework for obtaining targeted policies that restrict economic activity in different neighborhoods of a city at different levels. We focus on COVID-19 and calibrate our model using the mobile phone data that capture individuals’ movements within New York City (NYC). We use these data to illustrate that targeting can allow for substantially higher employment levels than uniform (city-wide) policies when applied to reduce infections across a region of focus. In our NYC example (which focuses on the control of the disease in April 2020), our main model illustrates that appropriate targeting achieves a reduction in infections in all neighborhoods while resuming 23.1%–42.4% of the baseline nonteleworkable employment level. By contrast, uniform restriction policies that achieve the same policy goal permit 3.92–6.25 times less nonteleworkable employment. Our optimization framework demonstrates the potential of targeting to limit the economic costs of unemployment while curbing the spread of an epidemic.

This paper was accepted by Carri Chan, healthcare management.

 

We consider the problem of controlling a Linear Quadratic Regulator (LQR) system over a finite horizon T with fixed and known cost matrices Q,R, but unknown and non-stationary dynamics A_t, B_t. The sequence of dynamics matrices can be arbitrary, but with a total variation, V_T, assumed to be o(T) and unknown to the controller. Under the assumption that a sequence of stabilizing, but potentially sub-optimal controllers is available for all t, we present an algorithm that achieves the optimal dynamic regret of O(V_T^2/5 T^3/5 ). With piecewise constant dynamics, our algorithm achieves the optimal regret of O(sqrtST ) where S is the number of switches. The crux of our algorithm is an adaptive non-stationarity detection strategy, which builds on an approach recently developed for contextual Multi-armed Bandit problems. We also argue that non-adaptive forgetting (e.g., restarting or using sliding window learning with a static window size) may not be regret optimal for the LQR problem, even when the window size is optimally tuned with the knowledge of $V_T$. The main technical challenge in the analysis of our algorithm is to prove that the ordinary least squares (OLS) estimator has a small bias when the parameter to be estimated is non-stationary. Our analysis also highlights that the key motif driving the regret is that the LQR problem is in spirit a bandit problem with linear feedback and locally quadratic cost. This motif is more universal than the LQR problem itself, and therefore we believe our results should find wider application.