The coronavirus pandemic has ushered in a new era of workplace surveillance that will extend well beyond our current crisis. Companies are increasingly monitoring employees who work from home, citing worries about security concerns or the need to boost employee productivity. In Amazon warehouses and UPS delivery trucks, surveillance technologies are being built into workplace infrastructure to monitor workers' every move. In many industries, employers can easily access phone calls, texts, browser histories, emails, and even GPS locations with very little effort. These exploitative surveillance practices are rooted in a historical mistrust of workers, especially low-wage workers, that can arguably be traced back to slavery and the exploitative "scientific management" practices that emerged from it, as bosses became obsessed with tracking workers' every move to maximize productivity and profit. Earlier forms of surveillance, like in the 19th century when companies hired Pinkerton private detectives to spy on workers, required a lot of labor. But modern technological advancements mean that the cost of surveillance today is very low.

Matrix completion has been well studied under the uniform sampling model and the trace-norm regularized methods perform well both theoretically and numerically in such a setting. However, the uniform sampling model is unrealistic for a range of applications and the standard trace-norm relaxation can behave very poorly when the underlying sampling scheme is non-uniform. In this paper we propose and analyze a max-norm constrained empirical risk minimization method for noisy matrix completion under a general sampling model. The optimal rate of convergence is established under the Frobenius norm loss in the context of approximately low-rank matrix reconstruction. It is shown that the max-norm constrained method is minimax rate-optimal and yields a unified and robust approximate recovery guarantee, with respect to the sampling distributions. The computational effectiveness of this method is also discussed, based on first-order algorithms for solving convex optimizations involving max-norm regularization.