In this paper, we introduce autoencoder ensembles for unsupervised outlier detection. One problem with neural networks is that they are sensitive to noise and often require large data sets to work robustly, while increasing data size makes them slow. As a result, there are only a few existing works in the literature on the use of neural networks in outlier detection. This paper shows that neural networks can be a very competitive technique to other existing methods. The basic idea is to randomly vary on the connectivity architecture of the autoencoder to obtain significantly better performance. Furthermore, we combine this technique with an adaptive sampling method to make our approach more efficient and effective. Experimental results comparing the proposed approach with state-of-theart detectors are presented on several benchmark data sets showing the accuracy of our approach.

Feature engineering is the task of improving predictive modelling performance on a dataset by transforming its feature space.Existing approaches to automate this process rely on either transformed feature space exploration through evaluation-guided search, or explicit expansion of datasets with all transformed features followed by feature selection. Such approaches incur high computational costs in runtime and/or memory. We present a novel technique, called Learning Feature Engineering (LFE), for automating feature engineering in classification tasks. LFE is based on learning the effectiveness of applying a transformation (e.g., arithmetic or aggregate operators) on numerical features, from past feature engineering experiences. Given a new dataset, LFE recommends a set of useful transformations to be applied on features without relying on model evaluation or explicit feature expansion and selection. Using a collection of datasets, we train a set of neural networks, which aim at predicting the transformation that impacts classification performance positively. Our empirical results show that LFE outperforms other feature engineering approaches for an overwhelming majority (89%) of the datasets from various sources while incurring a substantially lower computational cost.

Data augmentation is the process of generating samples by transforming training data, with the target of improving the accuracy and robustness of classifiers. In this paper, we propose a new automatic and adaptive algorithm for choosing the transformations of the samples used in data augmentation. Specifically, for each sample, our main idea is to seek a small transformation that yields maximal classification loss on the transformed sample. We employ a trust-region optimization strategy, which consists of solving a sequence of linear programs. Our data augmentation scheme is then integrated into a Stochastic Gradient Descent algorithm for training deep neural networks. We perform experiments on two datasets, and show that that the proposed scheme outperforms random data augmentation algorithms in terms of accuracy and robustness, while yielding comparable or superior results with respect to existing selective sampling approaches.

Existing wireless networks provide dynamically varying resources with only limited support for the Quality of Service required by the bandwidth-intense, loss-tolerant, and delay-sensitive multimedia applications. This variability of resources does not significantly impact data applications (e.g., file transfers), but has considerable consequences for multimedia applications and often leads to unsatisfactory user experience. Recently, the research focus has been to adapt existing algorithms and protocols at the lower layers of the network stack to better support multimedia transmission applications, and conversely, to modify application layer solutions to cope with the varying wireless networks resources. In this paper, we show that significant improvements in wireless multimedia performance can be obtained by deploying a joint application-layer packetization and MAC-layer retransmission strategy. First, we show that packet-size optimizations solely determined at the MAC-layer result in a sub-optimal performance in terms of the multimedia quality. Subsequently, we propose cross-layer strategies that optimize the packetization, prioritization and retransmission strategies based on content characteristics, channel conditions, and the specific features of the deployed video coder. Finally, we investigate the use of content based distortion models for the video, to reduce the complexity of our proposed optimization. 1. Introduction Wireless networks provide only limited support for the Quality of Service (QoS) required by delay-sensitive and high-bandwidth multimedia applications as they provide dynamically varying resources in terms of available bandwidth, due to multi-path fading, co-channel interference, and noise disturbances. A variety of application-layer solutions have been proposed to cope with these challenges. These include rate adaptation, (rate-distortion optimized) scheduling, error resilience techniques, error concealment mechanisms and joint source-channel coding. An excellent review of application-layer research in wireless multimedia streaming is provided in [1]. Cross-layer design for wireless multimedia transmission has also been investigated (e.g. [7][11][15]) and the results indicate that a significant gain in performance can be obtained. It is however important to note that existing cross-layer solutions often overlook the important issue of packetization and its relationship to other protection strategies at various layers as well as its impact on the ratedistortion (R-D) performance at the application-layer. In this paper, we focus on developing content-based flexible and adaptive packetization strategies for scalable multimedia streams and corresponding Medium Access Control (MAC) retransmission strategies to enable optimal rate-distortion-resilience tradeoffs for wireless multimedia streaming. We develop these joint packetization-retransmission schemes using a crosslayer optimization approach, where the application layer collaborates with the MAC layer to jointly determine the optimal packet ...

No abstract