In this paper, we design a navigation policy for multiple unmanned aerial vehicles (UAVs) where mobile base stations (BSs) are deployed to improve the data freshness and connectivity to the Internet of Things (IoT) devices. First, we formulate an energy-efficient trajectory optimization problem in which the objective is to maximize the energy efficiency by optimizing the UAV-BS trajectory policy. We also incorporate different contextual information such as energy and age of information (AoI) constraints to ensure the data freshness at the ground BS. Second, we propose an agile deep reinforcement learning with experience replay model to solve the formulated problem concerning the contextual constraints for the UAV-BS navigation. Moreover, the proposed approach is well-suited for solving the problem, since the state space of the problem is extremely large and finding the best trajectory policy with useful contextual features is too complex for the UAV-BSs. By applying the proposed trained model, an effective real-time trajectory policy for the UAV-BSs captures the observable network states over time. Finally, the simulation results illustrate the proposed approach is 3.6% and 3.13% more energy efficient than those of the greedy and baseline deep Q Network (DQN) approaches.

ABSTRACT different granularities [3, 9] or use a cross interaction block that couple the features from different modalities [10, 6]. It is imperative that all modalities in multimodal interactions and 3. Fusion of unimodal and cross Therefore, to learn better cross modal information, we introduce 1.6% and 1.34% absolute improvement over current state-ofthe-art. Furthermore, to capture long term dependencies across 1. INTRODUCTION These are categorised into three types, 1. Methods that learn the modalities independently and fuse the In our proposed model, we aim to learn the interaction between [3, 4], and 3. Methods that explicitly learn contributions Personal use of this material is permitted. Multimodal sentiment analysis provides an opportunity to 2.1. M T V H T W H T V; W R d d (3) (U 1, U 2,..., U u) for a Text modality can be defined as: Cross attentive representations of Text (C V T R u d) and H T Bi-GRU(U 1, U 2,..., U u) (1) Video (C T V R u d) can be represented as: Subscript T denotes Text modality, A and V represent Audio As much as there is an opportunity to leverage cross modal interactions, representations is employed.

Many state-of-the-art technologies developed in recent years have been influenced by machine learning to some extent. Most popular at the time of this writing are artificial intelligence methodologies that fall under the umbrella of deep learning. Deep learning has been shown across many applications to be extremely powerful and capable of handling problems that possess great complexity and difficulty. In this work, we introduce a new layer to deep learning: the fuzzy layer. Traditionally, the network architecture of neural networks is composed of an input layer, some combination of hidden layers, and an output layer. We propose the introduction of fuzzy layers into the deep learning architecture to exploit the powerful aggregation properties expressed through fuzzy methodologies, such as the Choquet and Sugueno fuzzy integrals. To date, fuzzy approaches taken to deep learning have been through the application of various fusion strategies at the decision level to aggregate outputs from state-of-the-art pre-trained models, e.g., AlexNet, VGG16, GoogLeNet, Inception-v3, ResNet-18, etc. While these strategies have been shown to improve accuracy performance for image classification tasks, none have explored the use of fuzzified intermediate, or hidden, layers. Herein, we present a new deep learning strategy that incorporates fuzzy strategies into the deep learning architecture focused on the application of semantic segmentation using per-pixel classification. Experiments are conducted on a benchmark data set as well as a data set collected via an unmanned aerial system at a U.S. Army test site for the task of automatic road segmentation, and preliminary results are promising.

In this paper we demonstrate speech synthesis using different electroencephalography (EEG) feature sets recently introduced in [1]. We make use of a recurrent neural network (RNN) regression model to predict acoustic features directly from EEG features. We demonstrate our results using EEG features recorded in parallel with spoken speech as well as using EEG recorded in parallel with listening utterances. We provide EEG based speech synthesis results for four subjects in this paper and our results demonstrate the feasibility of synthesizing speech directly from EEG features.

Music that is generated by recurrent neural networks often lacks a sense of direction and coherence. We therefore propose a two-stage LSTM-based model for lead sheet generation, in which the harmonic and rhythmic templates of the song are produced first, after which, in a second stage, a sequence of melody notes is generated conditioned on these templates. A subjective listening test shows that our approach outperforms the baselines and increases perceived musical coherence.

Nonconvex minimax problems appear frequently in emerging machine learning applications, such as generative adversarial networks and adversarial learning. Simple algorithms such as the gradient descent ascent (GDA) are the common practice for solving these nonconvex games and receive lots of empirical success. Yet, it is known that these vanilla GDA algorithms with constant step size can potentially diverge even in the convex setting. In this work, we show that for a subclass of nonconvex-nonconcave objectives satisfying a so-called two-sided Polyak-{\L}ojasiewicz inequality, the alternating gradient descent ascent (AGDA) algorithm converges globally at a linear rate and the stochastic AGDA achieves a sublinear rate. We further develop a variance reduced algorithm that attains a provably faster rate than AGDA when the problem has the finite-sum structure.

We consider the problem of estimating a ranking on a set of items from noisy pairwise comparisons given item features. We address the fact that pairwise comparison data often reflects irrational choice, e.g. intransitivity. Our key observation is that two items compared in isolation from other items may be compared based on only a salient subset of features. Formalizing this framework, we propose the "salient feature preference model" and prove a sample complexity result for learning the parameters of our model and the underlying ranking with maximum likelihood estimation. We also provide empirical results that support our theoretical bounds and illustrate how our model explains systematic intransitivity. Finally we demonstrate strong performance of maximum likelihood estimation of our model on both synthetic data and two real data sets: the UT Zappos50K data set and comparison data about the compactness of legislative districts in the US.

Modern machine learning systems often leverage data that are generated ubiquitously and seamlessly through devices such as smartphones, cameras, microphones, or user's weblogs, transaction logs, social media, etc. Much of this data is private, and releasing models trained on such data without serious privacy considerations can reveal sensitive information (Narayanan and Shmatikov, 2008; Sweeney, 1997). Consequently, much emphasis has been placed in recent years on machine learning under the constraints of a robust privacy guarantee. One such notion that has emerged as a de facto standard is that of differential privacy. Informally, differential privacy provides a quantitative assessment of how different are the outputs of a randomized algorithm when fed two very similar inputs. If small changes in the input do not manifest as drastically different outputs, then it is hard to discern much information about the inputs solely based on the outputs of the algorithm. In the context of machine learning, this implies that if the learning algorithm is not overly sensitive to any single datum in the training set, then releasing the trained model should preserve the privacy of the training data. This requirement, apriori, seems compatible with the goal of learning, which is to find a model that generalizes well on the population and does not overfit to the given training sample. It seems reasonable then to argue that privacy is not necessarily at odds with generalization, especially when large training sets are available.

Nowadays, graph-structured data are increasingly used to model complex systems. Meanwhile, detecting anomalies from graph has become a vital research problem of pressing societal concerns. Anomaly detection is an unsupervised learning task of identifying rare data that differ from the majority. As one of the dominant anomaly detection algorithms, One Class Support Vector Machine has been widely used to detect outliers. However, those traditional anomaly detection methods lost their effectiveness in graph data. Since traditional anomaly detection methods are stable, robust and easy to use, it is vitally important to generalize them to graph data. In this work, we propose One Class Graph Neural Network (OCGNN), a one-class classification framework for graph anomaly detection. OCGNN is designed to combine the powerful representation ability of Graph Neural Networks along with the classical one-class objective. Compared with other baselines, OCGNN achieves significant improvements in extensive experiments.

Sample- and computationally-efficient distribution estimation is a fundamental tenet in statistics and machine learning. We present $\mathrm{SURF}$, an algorithm for approximating distributions by piecewise polynomials. $\mathrm{SURF}$ is simple, replacing existing general-purpose optimization techniques by straight-forward approximation of each potential polynomial piece by a simple empirical-probability interpolation, and using plain divide-and-conquer to merge the pieces. It is universal, as well-known low-degree polynomial-approximation results imply that it accurately approximates a large class of common distributions. $\mathrm{SURF}$ is robust to distribution mis-specification as for any degree $d\le 8$, it estimates any distribution to an $\ell_1$ distance $ <3 $ times that of the nearest degree-$d$ piecewise polynomial, improving known factor upper bounds of 3 for single polynomials and 15 for polynomials with arbitrarily many pieces. It is fast, using optimal sample complexity, and running in near sample-linear time. In experiments, $\mathrm{SURF}$ significantly outperforms state-of-the art algorithms.