A Generative Neighborhood-Based Deep Autoencoder for Robust Imbalanced Classification.

Deep learning models perform remarkably well on many classification tasks recently. The superior performance of deep neural networks relies on the large number of training data, which at the same time must have an equal class distribution in order to be efficient. However, in most real-world applications, the labeled data may be limited with high imbalance ratios among the classes, and thus, the learning process of most classification algorithms is adversely affected resulting in unstable predictions and low performance. Three main categories of approaches address the problem of imbalanced learning, i.e., data-level, algorithmic level, and hybrid methods, which combine the two aforementioned approaches. Data generative methods are typically based on generative adversarial networks, which require significant amounts of data, while model-level methods entail extensive domain expert knowledge to craft the learning objectives, thereby being less accessible for users without such knowledge. Moreover, the vast majority of these approaches are designed and applied to imaging applications, less to time series, and extremely rare to both of them. To address the above issues, we introduce GENDA, a generative neighborhood-based deep autoencoder, which is simple yet effective in its design and can be successfully applied to both image and time-series data. GENDA is based on learning latent representations that rely on the neighboring embedding space of the samples. Extensive experiments, conducted on a variety of widely-used real datasets demonstrate the efficacy of the proposed method. Impact Statement­: Imbalanced data classification is an actual and important issue in many real-world learning applications hampering most classification tasks. Fraud detection, biomedical imaging categorizing healthy people versus patients, and object detection are some indicative domains with an economic, social and technological impact, which are greatly affected by inherent imbalanced data distribution. However, the majority of the existing algorithms that address the imbalanced classification problem are designed with a particular application in mind, and thus they can be used with specific datasets and even hyperparameters. The generative model introduced in this paper overcomes this limitation and produces improved results for a large class of imaging and time series data even under severe imbalance ratios, making it quite competitive.

Integrating Visual and Network Data with Deep Learning for Streaming Video Quality Assessment.

Existing video Quality-of-Experience (QoE) metrics rely on the decoded video for the estimation. In this work, we explore how the overall viewer experience, quantified via the QoE score, can be automatically derived using only information available before and during the transmission of videos, on the server side. To validate the merits of the proposed scheme, we consider a dataset of videos encoded and streamed under different conditions and train a novel deep learning architecture for estimating the QoE of the decoded video. The major novelty of our work is the exploitation and demonstration of cutting-edge deep learning techniques in automatically estimating video QoE scores. Our work significantly extends the existing approach for estimating the QoE in video streaming services by combining visual information and network conditions.

Embedded Temporal Convolutional Networks for Essential Climate Variables Forecasting.

Forecasting the values of essential climate variables like land surface temperature and soil moisture can play a paramount role in understanding and predicting the impact of climate change. This work concerns the development of a deep learning model for analyzing and predicting spatial time series, considering both satellite derived and model-based data assimilation processes. To that end, we propose the Embedded Temporal Convolutional Network (E-TCN) architecture, which integrates three different networks, namely an encoder network, a temporal convolutional network, and a decoder network. The model accepts as input satellite or assimilation model derived values, such as land surface temperature and soil moisture, with monthly periodicity, going back more than fifteen years. We use our model and compare its results with the state-of-the-art model for spatiotemporal data, the ConvLSTM model. To quantify performance, we explore different cases of spatial resolution, spatial region extension, number of training examples and prediction windows, among others. The proposed approach achieves better performance in terms of prediction accuracy, while using a smaller number of parameters compared to the ConvLSTM model. Although we focus on two specific environmental variables, the method can be readily applied to other variables of interest.

Large-Scale 3D Two-Photon Imaging of Molecularly Identified CA1 Interneuron Dynamics in Behaving Mice.

Cortical computations are critically reliant on their local circuit, GABAergic cells. In the hippocampus, a large body of work has identified an unprecedented diversity of GABAergic interneurons with pronounced anatomical, molecular, and physiological differences. Yet little is known about the functional properties and activity dynamics of the major hippocampal interneuron classes in behaving animals. Here we use fast, targeted, three-dimensional (3D) two-photon calcium imaging coupled with immunohistochemistry-based molecular identification to retrospectively map in vivo activity onto multiple classes of interneurons in the mouse hippocampal area CA1 during head-fixed exploration and goal-directed learning. We find examples of preferential subtype recruitment with quantitative differences in response properties and feature selectivity during key behavioral tasks and states. These results provide new insights into the collective organization of local inhibitory circuits supporting navigational and mnemonic functions of the hippocampus.

Survey of Deep-Learning Approaches for Remote Sensing Observation Enhancement.

Deep Learning, and Deep Neural Networks in particular, have established themselves as the new norm in signal and data processing, achieving state-of-the-art performance in image, audio, and natural language understanding. In remote sensing, a large body of research has been devoted to the application of deep learning for typical supervised learning tasks such as classification. Less yet equally important effort has also been allocated to addressing the challenges associated with the enhancement of low-quality observations from remote sensing platforms. Addressing such channels is of paramount importance, both in itself, since high-altitude imaging, environmental conditions, and imaging systems trade-offs lead to low-quality observation, as well as to facilitate subsequent analysis, such as classification and detection. In this paper, we provide a comprehensive review of deep-learning methods for the enhancement of remote sensing observations, focusing on critical tasks including single and multi-band super-resolution, denoising, restoration, pan-sharpening, and fusion, among others. In addition to the detailed analysis and comparison of recently presented approaches, different research avenues which could be explored in the future are also discussed.

Matrix and Tensor Completion on a Human Activity Recognition Framework.

Sensor-based activity recognition is encountered in innumerable applications of the arena of pervasive healthcare and plays a crucial role in biomedical research. Nonetheless, the frequent situation of unobserved measurements impairs the ability of machine learning algorithms to efficiently extract context from raw streams of data. In this paper, we study the problem of accurate estimation of missing multimodal inertial data and we propose a classification framework that considers the reconstruction of subsampled data during the test phase. We introduce the concept of forming the available data streams into low-rank two-dimensional (2-D) and 3-D Hankel structures, and we exploit data redundancies using sophisticated imputation techniques, namely matrix and tensor completion. Moreover, we examine the impact of reconstruction on the classification performance by experimenting with several state-of-the-art classifiers. The system is evaluated with respect to different data structuring scenarios, the volume of data available for reconstruction, and various levels of missing values per device. Finally, the tradeoff between subsampling accuracy and energy conservation in wearable platforms is examined. Our analysis relies on two public datasets containing inertial data, which extend to numerous activities, multiple sensing parameters, and body locations. The results highlight that robust classification accuracy can be achieved through recovery, even for extremely subsampled data streams.

Dictionary Learning for Spontaneous Neural Activity Modeling.

Modeling the activity of an ensemble of neurons can provide critical insights into the workings of the brain. In this work we examine if learning based signal modeling can contribute to a high quality modeling of neuronal signal data. To that end, we employ the sparse coding and dictionary learning schemes for capturing the behavior of neuronal responses into a small number of representative prototypical signals. Performance is measured by the reconstruction quality of clean and noisy test signals, which serves as an indicator of the generalization and discrimination capabilities of the learned dictionaries. To validate the merits of the proposed approach, a novel dataset of the actual recordings from 183 neurons from the primary visual cortex of a mouse in early postnatal development was developed and investigated. The results demonstrate that high quality modeling of testing data can be achieved from a small number of training examples and that the learned dictionaries exhibit significant specificity when introducing noise.