Search

Several recent papers have discussed utilizing Lipschitz constants to limit the susceptibility of neural networks to adversarial examples. We analyze recently proposed methods for computing the Lipschitz constant. We show that the Lipschitz constant may indeed enable adversarially robust neural networks. However, the methods currently employed for computing it suffer from theoretical and practical limitations. We argue that addressing this shortcoming is a promising direction for future research into certified adversarial defenses.

Testing of applications for tactical MANETs poses a special technical challenge due to the difficulty of conducting experiments in an ad hoc network environment at a scale larger than a few nodes. One approach is to conduct experiments in Virtual Ad hoc Network (VAN) testbeds1 that can imitate a tactical MANET to a high level of fidelity. For applications, this is achieved by executing unmodified software in their native operating systems on the machines over an emulated platform that uses identical logical layout of the real physical platform; for the network, it is achieved by using an emulated/simulated network that behaves like a real MANET in some/all layers of the protocol stack. In this paper, we analyze the challenges and tradeoffs of realistic modeling of tactical networks with multilevel security (MLS), and present a case study of modeling BCTM-like manned ground vehicle (MGV) platforms in the VAN Testbed. Lessons learned during application development and testing performed on the model are also discussed.

Deadlock detection scheduling is an important, yet often overlooked problem that can significantly affect the overall performance of deadlock handling. Excessive initiation of deadlock detection increases overall message usage, resulting in degraded system performance in the absence of deadlocks; while insufficient initiation of deadlock detection increases the deadlock persistence time, resulting in an increased deadlock resolution cost in the presence of deadlocks. The investigation of this performance tradeoff, however, is missing in the literature. This paper studies the impact of deadlock detection scheduling on the overall performance of deadlock handling. In particular, we show that there exists an optimal deadlock detection frequency that yields the minimum long-run mean average cost, which is determined by the message complexities of the deadlock detection and resolution algorithms being used, as well as the rate of deadlock formation, denoted as $\lambda$. For the best known deadlock detection and resolution algorithms, we show that the asymptotically optimal frequency of deadlock detection scheduling that minimizes the overall message overhead is ${\cal O}((\lambda n)^{1/3})$, when the total number $n$ of processes is sufficiently large. Furthermore, we show that in general fully distributed (uncoordinated) deadlock detection scheduling cannot be performed as efficiently as centralized (coordinated) deadlock detection scheduling.