We provide a general and easy to implement method for reducing the number of parameters of Convolutional Neural Networks (CNNs) during the training and inference phases. We introduce a simple trainable auxiliary neural network which can generate approximate versions of “slices” of the sets of convolutional filters of any CNN architecture from a low dimensional “code” space. These slices are then concatenated to form the sets of filters in the CNN architecture. The auxiliary neural network, which we call “Convolutional Slice Generator” (CSG), is unique to the network and provides the association among its convolutional layers. We apply our method to various CNN architectures including ResNet, DenseNet, MobileNet and ShuffleNet. Experiments on CIFAR-10 and ImageNet-1000, without any hyper-parameter tuning, show that our approach reduces the network parameters by approximately $2\times$ while the reduction in accuracy is confined to within one percent and sometimes the accuracy even improves after compression. Interestingly, through our experiments, we show that even when the CSG takes random binary values for its weights that are not learned, still acceptable performances are achieved. To show that our approach generalizes to other tasks, we apply it to an image segmentation architecture, Deeplab V3, on the Pascal VOC 2012 dataset. Results show that without any parameter tuning, there is $\approx 2.3\times$ parameter reduction and the mean Intersection over Union (mIoU) drops by $\approx 3%$. Finally, we provide comparisons with several related methods showing the superiority of our method in terms of accuracy.

We consider an agent-based distributed algorithm with exponentially distributed waiting times in which agents with binary states interact locally over a geometric graph, and based on this interaction and on the value of a common intolerance threshold τ, decide whether to change their states. This model is equivalent to an Asynchronous Cellular Automaton (ACA) with extended Moore neighborhoods, a zero-temperature Ising model with Glauber dynamics, or a Schelling model of self-organized segregation in an open system, and has applications in the analysis of social and biological networks, and spin glasses systems.

We consider an agent-based model in which two types of agents interact locally over a graph and have a common intolerance threshold τ for changing their types with exponentially distributed waiting times. The model is equivalent to an unperturbed Schelling model of self-organized segregation, an Asynchronous Cellular Automata (ACA) with extended Moore neighborhoods, or a zero-temperature Ising model with Glauber dynamics, and has applications in the analysis of social and biological networks, and spin glasses systems. Some rigorous results were recently obtained in the theoretical computer science literature, and this work provides several extensions. We enlarge the intolerance interval leading to the formation of large segregated regions of agents of a single type from the known size ε>0 to size ~0.134. Namely, we show that for 0.433 < τ < 1/2 (and by symmetry 1/2<τ<0.567), the expected size of the largest segregated region containing an arbitrary agent is exponential in the size of the neighborhood. We further extend the interval leading to large segregated regions to size ~0.312 considering "almost segregated" regions, namely regions where the ratio of the number of agents of one type and the number of agents of the other type vanishes quickly as the size of the neighborhood grows. In this case, we show that for 0.344 < τ ≤ 0.433 (and by symmetry for 0.567 ≤ τ<0.656) the expected size of the largest almost segregated region containing an arbitrary agent is exponential in the size of the neighborhood. This behavior is reminiscent of supercritical percolation, where small clusters of empty sites can be observed within any sufficiently large region of the occupied percolation cluster. The exponential bounds that we provide also imply that complete segregation, where agents of a single type cover the whole grid, does not occur with high probability for p=1/2 and the range of tolerance considered.

In wireless body area networks (WBANs) it is necessary to devise an energy-efficient MAC scheduling mechanism which is capable of meeting strict QoS requirements. In this paper, special characteristics of WBAN channels, namely, the slow fading, the periodicity of fading, and the correlation among the channels have been exploited to formulate the sensor scheduling problem as a partially observable Markov decision problem (POMDP). Specific algorithms based on value iteration and pruning have been proposed in order to reduce the computational burden associated with the corresponding POMDP. The proposed scheduling mechanism is compared against a TDMA scheduling for a three-sensor network, and an average 22-32% improvement in energy consumption and slight improvement in reliability is reported.