This work concerns the development of deep networks that are certifiably robust to adversarial attacks. Joint robust classification-detection was recently introduced as a certified defense mechanism, where adversarial examples are either correctly classified or assigned to the “abstain” class. In this work, we show that such a provable framework can benefit by extension to networks with multiple explicit abstain classes, where the adversarial examples are adaptively assigned to those. We show that naïvely adding multiple abstain classes can lead to “model degeneracy”, then we propose a regularization approach and a training method to counter this degeneracy by promoting full use of the multiple abstain classes. Our experiments demonstrate that the proposed approach consistently achieves favorable standard vs. robust verified accuracy tradeoffs, outperforming state-of-the-art algorithms for various choices of number of abstain classes.

Despite their well-documented learning capabilities in clean environments, deep convolutional neural networks (CNNs) are extremely fragile in adversarial settings, where carefully crafted perturbations created by an attacker can easily disrupt the task at hand. Numerous methods have been proposed for designing effective attacks, while the design of effective defense schemes is still an open area. This work leverages randomization-based defense schemes to introduce a sampling mechanism for strong and efficient defense. To this end, sampling is proposed to take place over the matricized mid-layer data in the neural network, and the sampling probabilities are systematically obtained via variance minimization. The proposed defense only requires adding sampling blocks to the network in the inference phase without extra overhead in the training. In addition, it can be utilized on any pre-trained network without altering the weights. Numerical tests corroborate the improved defense against various attack schemes in comparison with state-of-the-art randomized defenses.

Small base stations (SBs) of fifth-generation (5G) cellular networks are envisioned to have storage devices to locally serve requests for reusable and popular contents by caching them at the edge of the network, close to the end users. The ultimate goal is to smartly utilize a limited storage capacity to serve locally contents that are frequently requested instead of fetching them from the cloud, contributing to a better overall network performance and service experience. To enable the SBs with efficient fetch-cache decision-making schemes operating in dynamic settings, this paper introduces simple but flexible generic time-varying fetching and caching costs, which are then used to formulate a constrained minimization of the aggregate cost across files and time. Since caching decisions per time slot influence the content availability in future slots, the novel formulation for optimal fetch-cache decisions falls into the class of dynamic programming. Under this generic formulation, first by considering stationary distributions for the costs as well as file popularities, an efficient reinforcement learning-based solver known as value iteration algorithm can be used to solve the emerging optimization problem. Later, it is shown that practical limitations on cache capacity can be handled using a particular instance of this generic dynamic pricing formulation. Under this setting, to provide a light-weight online solver for the corresponding optimization, the well-known reinforcement learning algorithm, Q-learning, is employed to find optimal fetch-cache decisions. Numerical tests corroborating the merits of the proposed approach wrap up the paper.

In next generation cellular networks (5G) the access points (APs) are anticipated to be equipped with storage devices to serve locally requests for reusable popular contents by caching them at the edge of the network. The ultimate goal is to shift part of the load on the back-haul links from on-peak to off-peak periods, contributing to a better overall network performance and service experience. In order to enable the APs with efficient (optimal) fetch-cache decision making schemes able to work in dynamic settings, we introduce simple but flexible generic time-varying fetching and caching costs, which are then used to formulate a constrained minimization of the aggregate cost across files and time. Since caching decisions in every time slot influence the content availability in future instants, the novel formulation for optimal fetch-cache decisions falls into the class of dynamic programming, for which efficient reinforcement-learning-based solvers are proposed. The performance of our algorithms is assessed via numerical tests, and discussions on the inherent fetching-versus-caching trade-off are provided.

The task of community detection over complex networks is of paramount importance in a multitude of applications. The present work puts forward a top-to-bottom community identification approach, termed DC-EgoTen, in which an egonet-tensor (EgoTen) based algorithm is developed in a divide-and-conquer (DC) fashion for breaking the network into smaller subgraphs, out of which the underlying communities progressively emerge. In particular, each step of DC-EgoTen forms a multi-dimensional egonet-based representation of the graph, whose induced structure enables casting the task of overlapping community identification as a constrained PARAFAC decomposition. Thanks to the higher representational capacity of tensors, the novel egonet-based representation improves the quality of detected communities by capturing multi-hop connectivity patterns of the network. In addition, the top-to-bottom approach ensures successive refinement of identified communities, so that the desired resolution is achieved. Synthetic as well as real-world tests corroborate the effectiveness of DC-EgoTen.

Spectrum mobility in cognitive radio networks not only enables the secondary users to guarantee the desired QoS of the primary users but also grants an efficient exploitation of the available spectrum holes in the network. In this paper, we propose a probabilistic approach in determining the initial and target channels for the handoff procedure in a single secondary user network. To characterize the network, a queuing theoretical framework is introduced, and “stay” and “change” handoff policies are both addressed. The performance of the secondary user in terms of average sojourn and extended service times for secondary connections is analyzed, and convex optimization problems with the objective of minimizing those times as well as analytical solutions are presented. Simulation results confirm the validity of our analytical approach.

Effective spectrum sensing strategies enable cognitive radios (CRs) to identify and opportunistically transmit on under-utilized spectral resources. In this paper, sequential channel sensing problems for single and multiple secondary users (SUs) networks are effectively modeled through finite state Markovian processes. More specifically, a model for single user case is introduced and its performance is validated through analytical analysis. Then, in order to address multiple SUs case, this model is extended to include the modified p-persistent access (MPPA) protocol. Since the scheme utilized experiences a high level of collision among the SUs, to mitigate the problem appropriately, p-persistent random access (PPRA) protocol is considered, which offers higher average throughput for SUs by statistically distributing their loads among all channels. The structure and performance of the proposed schemes are discussed in detail, and a set of illustrative numerical results is presented to validate and compare the performance of the proposed sense-access strategies.