Sequence-to-Sequence-to-Sequence Autoencoder (SEQ^3)
Neural sequence-to-sequence models are currently the dominant approach in several natural language processing tasks, but require large parallel corpora. We present a sequence-to-sequence-to-sequence autoencoder (SEQ^3), consisting of two chained encoder-decoder pairs, with words used as a sequence of discrete latent variables. We apply the proposed model to unsupervised abstractive sentence compression, where the first and last sequences are the input and reconstructed sentences, respectively, while the middle sequence is the compressed sentence. Constraining the length of the latent word sequences forces the model to distill important information from the input. A pretrained language model, acting as a prior over the latent sequences, encourages the compressed sentences to be human-readable. Continuous relaxations enable us to sample from categorical distributions, allowing gradient-based optimization, unlike alternatives that rely on reinforcement learning. The proposed model does not require parallel text-summary pairs, achieving promising results in unsupervised sentence compression on benchmark datasets. …

Active Domain Randomization
Domain randomization is a popular technique for improving domain transfer, often used in a zero-shot setting when the target domain is unknown or cannot easily be used for training. In this work, we empirically examine the effects of domain randomization on agent generalization. Our experiments show that domain randomization may lead to suboptimal, high-variance policies, which we attribute to the uniform sampling of environment parameters. We propose Active Domain Randomization, a novel algorithm that learns a parameter sampling strategy. Our method looks for the most informative environment variations within the given randomization ranges by leveraging the discrepancies of policy rollouts in randomized and reference environment instances. We find that training more frequently on these instances leads to better overall agent generalization. In addition, when domain randomization and policy transfer fail, Active Domain Randomization offers more insight into the deficiencies of both the chosen parameter ranges and the learned policy, allowing for more focused debugging. Our experiments across various physics-based simulated and a real-robot task show that this enhancement leads to more robust, consistent policies. …

Advanced Supervised Principal Component Analysis (ASPCA)
We present a straightforward non-iterative method for shallowing of deep Convolutional Neural Network (CNN) by combination of several layers of CNNs with Advanced Supervised Principal Component Analysis (ASPCA) of their outputs. We tested this new method on a practically important case of `friend-or-foe’ face recognition. This is the backyard dog problem: the dog should (i) distinguish the members of the family from possible strangers and (ii) identify the members of the family. Our experiments revealed that the method is capable of drastically reducing the depth of deep learning CNNs, albeit at the cost of mild performance deterioration. …

Deep Multimodality Model for Multi-task Multi-view Learning (Deep-MTMV)
Many real-world problems exhibit the coexistence of multiple types of heterogeneity, such as view heterogeneity (i.e., multi-view property) and task heterogeneity (i.e., multi-task property). For example, in an image classification problem containing multiple poses of the same object, each pose can be considered as one view, and the detection of each type of object can be treated as one task. Furthermore, in some problems, the data type of multiple views might be different. In a web classification problem, for instance, we might be provided an image and text mixed data set, where the web pages are characterized by both images and texts. A common strategy to solve this kind of problem is to leverage the consistency of views and the relatedness of tasks to build the prediction model. In the context of deep neural network, multi-task relatedness is usually realized by grouping tasks at each layer, while multi-view consistency is usually enforced by finding the maximal correlation coefficient between views. However, there is no existing deep learning algorithm that jointly models task and view dual heterogeneity, particularly for a data set with multiple modalities (text and image mixed data set or text and video mixed data set, etc.). In this paper, we bridge this gap by proposing a deep multi-task multi-view learning framework that learns a deep representation for such dual-heterogeneity problems. Empirical studies on multiple real-world data sets demonstrate the effectiveness of our proposed Deep-MTMV algorithm. …

High Quality Bidirectional Generative Adversarial Network
Generative adversarial networks (GANs) have achieved outstanding success in generating the high quality data. Focusing on the generation process, existing GANs investigate unidirectional mapping from the latent vector to the data. Later, various studies point out that the latent space of GANs is semantically meaningful and can be utilized in advanced data analysis and manipulation. In order to analyze the real data in the latent space of GANs, it is necessary to investigate the inverse generation mapping from the data to the latent vector. To tackle this problem, the bidirectional generative models introduce an encoder to enable the inverse path of generation process. Unfortunately, this effort leads to the degradation of generation quality because the imperfect generator rather interferes the encoder training and vice versa. In this paper, we propose a new inference model that estimates the latent vector from the feature of GAN discriminator. While existing bidirectional models learns the image to latent translation, our algorithm formulates this inference mapping by the feature to latent translation. It is important to note that training of our model is independent of the GAN training. Owing to the attractive nature of this independency, the proposed algorithm can generate the high quality samples identical to those of unidirectional GANs and also reconstruct the original data faithfully. Moreover, our algorithm can be employed to any unidirectional GAN, even the pre-traind GANs. …

Generative Determinantal Point Process (GDPP)
Generative models have proven to be an outstanding tool for representing high-dimensional probability distributions and generating realistic looking images. A fundamental characteristic of generative models is their ability to produce multi-modal outputs. However, while training, they are often susceptible to mode collapse, which means that the model is limited in mapping the input noise to only a few modes of the true data distribution. In this paper, we draw inspiration from Determinantal Point Process (DPP) to devise a generative model that alleviates mode collapse while producing higher quality samples. DPP is an elegant probabilistic measure used to model negative correlations within a subset and hence quantify its diversity. We use DPP kernel to model the diversity in real data as well as in synthetic data. Then, we devise a generation penalty term that encourages the generator to synthesize data with a similar diversity to real data. In contrast to previous state-of-the-art generative models that tend to use additional trainable parameters or complex training paradigms, our method does not change the original training scheme. Embedded in an adversarial training and variational autoencoder, our Generative Determinantal Point Process approach shows a consistent resistance to mode-collapse on a wide-variety of synthetic data and natural image datasets including MNIST, CIFAR10, and CelebA, while outperforming state-of-the-art methods for data-efficiency, convergence-time, and generation quality. Our code is publicly available. …

Competitive Gradient Descent
We introduce a new algorithm for the numerical computation of Nash equilibria of competitive two-player games. Our method is a natural generalization of gradient descent to the two-player setting where the update is given by the Nash equilibrium of a regularized bilinear local approximation of the underlying game. It avoids oscillatory and divergent behaviors seen in alternating gradient descent. Using numerical experiments and rigorous analysis, we provide a detailed comparison to methods based on \emph{optimism} and \emph{consensus} and show that our method avoids making any unnecessary changes to the gradient dynamics while achieving exponential (local) convergence for (locally) convex-concave zero sum games. Convergence and stability properties of our method are robust to strong interactions between the players, without adapting the stepsize, which is not the case with previous methods. In our numerical experiments on non-convex-concave problems, existing methods are prone to divergence and instability due to their sensitivity to interactions among the players, whereas we never observe divergence of our algorithm. The ability to choose larger stepsizes furthermore allows our algorithm to achieve faster convergence, as measured by the number of model evaluations. …

Deep Bayesian Regression Model (DBRM)
Regression models are used for inference and prediction in a wide range of applications providing a powerful scientific tool for researchers and analysts from different fields. In many research fields the amount of available data as well as the number of potential explanatory variables is rapidly increasing. Variable selection and model averaging have become extremely important tools for improving inference and prediction. However, often linear models are not sufficient and the complex relationship between input variables and a response is better described by introducing non-linearities and complex functional interactions. Deep learning models have been extremely successful in terms of prediction although they are often difficult to specify and potentially suffer from overfitting. The aim of this paper is to bring the ideas of deep learning into a statistical framework which yields more parsimonious models and allows to quantify model uncertainty. To this end we introduce the class of deep Bayesian regression models (DBRM) consisting of a generalized linear model combined with a comprehensive non-linear feature space, where non-linear features are generated just like in deep learning but combined with variable selection in order to include only important features. DBRM can easily be extended to include latent Gaussian variables to model complex correlation structures between observations, which seems to be not easily possible with existing deep learning approaches. Two different algorithms based on MCMC are introduced to fit DBRM and to perform Bayesian inference. The predictive performance of these algorithms is compared with a large number of state of the art algorithms. Furthermore we illustrate how DBRM can be used for model inference in various applications. …

Prediction Advantage (PA)
We introduce the Prediction Advantage (PA), a novel performance measure for prediction functions under any loss function (e.g., classification or regression). The PA is defined as the performance advantage relative to the Bayesian risk restricted to knowing only the distribution of the labels. We derive the PA for well-known loss functions, including 0/1 loss, cross-entropy loss, absolute loss, and squared loss. In the latter case, the PA is identical to the well-known R-squared measure, widely used in statistics. The use of the PA ensures meaningful quantification of prediction performance, which is not guaranteed, for example, when dealing with noisy imbalanced classification problems. We argue that among several known alternative performance measures, PA is the best (and only) quantity ensuring meaningfulness for all noise and imbalance levels. …

Multi-Cell LSTM
Language models, being at the heart of many NLP problems, are always of great interest to researchers. Neural language models come with the advantage of distributed representations and long range contexts. With its particular dynamics that allow the cycling of information within the network, `Recurrent neural network’ (RNN) becomes an ideal paradigm for neural language modeling. Long Short-Term Memory (LSTM) architecture solves the inadequacies of the standard RNN in modeling long-range contexts. In spite of a plethora of RNN variants, possibility to add multiple memory cells in LSTM nodes was seldom explored. Here we propose a multi-cell node architecture for LSTMs and study its applicability for neural language modeling. The proposed multi-cell LSTM language models outperform the state-of-the-art results on well-known Penn Treebank (PTB) setup. …

Object Oriented Data Analysis (OODA)
Object oriented data analysis (OODA) is the statistical analysis of data sets of complex objects. The area is understood through consideration of the atom of the statistical analysis. In a first course in statistics, the atoms are numbers. Atoms are vectors in multivariate analysis. An interesting special case of OODA is functional data analysis, where atoms are curves; see Ramsay and Silverman for excellent overviews, as well as many interesting analyses, novel methodologies and detailed discussion. More general atoms have also been considered. Locantore et al. studied the case of images as atoms, and Pizer, Thall and Chen and Yushkevich et al. took the atoms to be shape objects in two- and three-dimensional space. An important major goal of OODA is understanding population structure of a data set. The usual first step is to find a centerpoint, for example, a mean or median, of the data set. The second step is to analyze the variation about the center. Principal component analysis (PCA) has been a workhorse method for this, especially when combined with new visualizations as done in functional data analysis. An important reason for this success to date is that the data naturally lie in Euclidean spaces, where standard vector space analyses have proven to be both insightful and effective. …

GaterNet
The concept of conditional computation for deep nets has been proposed previously to improve model performance by selectively using only parts of the model conditioned on the sample it is processing. In this paper, we investigate input-dependent dynamic filter selection in deep convolutional neural networks (CNNs). The problem is interesting because the idea of forcing different parts of the model to learn from different types of samples may help us acquire better filters in CNNs, improve the model generalization performance and potentially increase the interpretability of model behavior. We propose a novel yet simple framework called GaterNet, which involves a backbone and a gater network. The backbone network is a regular CNN that performs the major computation needed for making a prediction, while a global gater network is introduced to generate binary gates for selectively activating filters in the backbone network based on each input. Extensive experiments on CIFAR and ImageNet datasets show that our models consistently outperform the original models with a large margin. On CIFAR-10, our model also improves upon state-of-the-art results. …

Klout Score
Klout is a website and mobile app that uses social media analytics to rank its users according to online social influence via the ‘Klout Score’, which is a numerical value between 1 and 100. In determining the user score, Klout measures the size of a user’s social media network and correlates the content created to measure how other users interact with that content.
Klout Score: Measuring Influence Across Multiple Social Networks …

Neural Graph Collaborative Filtering (NGCF)
Learning vector representations (aka. embeddings) of users and items lies at the core of modern recommender systems. Ranging from early matrix factorization to recently emerged deep learning based methods, existing efforts typically obtain a user’s (or an item’s) embedding by mapping from pre-existing features that describe the user (or the item), such as ID and attributes. We argue that an inherent drawback of such methods is that, the collaborative signal, which is latent in user-item interactions, is not encoded in the embedding process. As such, the resultant embeddings may not be sufficient to capture the collaborative filtering effect. In this work, we propose to integrate the user-item interactions — more specifically the bipartite graph structure — into the embedding process. We develop a new recommendation framework Neural Graph Collaborative Filtering (NGCF), which exploits the user-item graph structure by propagating embeddings on it. This leads to the expressive modeling of high-order connectivity in user-item graph, effectively injecting the collaborative signal into the embedding process in an explicit manner. We conduct extensive experiments on three public benchmarks, demonstrating significant improvements over several state-of-the-art models like HOP-Rec and Collaborative Memory Network. Further analysis verifies the importance of embedding propagation for learning better user and item representations, justifying the rationality and effectiveness of NGCF. Codes are available at https://…/neural_graph_collaborative_filtering. …

Unremarkable AI
Clinical decision support tools (DST) promise improved healthcare outcomes by offering data-driven insights. While effective in lab settings, almost all DSTs have failed in practice. Empirical research diagnosed poor contextual fit as the cause. This paper describes the design and field evaluation of a radically new form of DST. It automatically generates slides for clinicians’ decision meetings with subtly embedded machine prognostics. This design took inspiration from the notion of ‘Unremarkable Computing’, that by augmenting the users’ routines technology/AI can have significant importance for the users yet remain unobtrusive. Our field evaluation suggests clinicians are more likely to encounter and embrace such a DST. Drawing on their responses, we discuss the importance and intricacies of finding the right level of unremarkableness in DST design, and share lessons learned in prototyping critical AI systems as a situated experience. …

Temporal Term Histogram (TTH)
Temporal text, i.e., time-stamped text data are found abundantly in a variety of data sources like newspapers, blogs and social media posts. While today’s data management systems provide facilities for searching full-text data, they do not provide any simple primitives for performing analytical operations with text. This paper proposes the temporal term histograms (TTH) as an intermediate primitive that can be used for analytical tasks. We propose an algebra, with operators and equivalence rules for TTH and present a reference implementation on a relational database system. …

Model Independent Neural Decoder (MIND)
Standard decoding approaches rely on model-based channel estimation methods to compensate for varying channel effects, which degrade in performance whenever there is a model mismatch. Recently proposed Deep learning based neural decoders address this problem by leveraging a model-free approach via gradient-based training. However, they require large amounts of data to retrain to achieve the desired adaptivity, which becomes intractable in practical systems. In this paper, we propose a new decoder: Model Independent Neural Decoder (MIND), which builds on the top of neural decoders and equips them with a fast adaptation capability to varying channels. This feature is achieved via the methodology of Model-Agnostic Meta-Learning (MAML). Here the decoder: (a) learns a ‘good’ parameter initialization in the meta-training stage where the model is exposed to a set of archetypal channels and (b) updates the parameter with respect to the observed channel in the meta-testing phase using minimal adaptation data and pilot bits. Building on top of existing state-of-the-art neural Convolutional and Turbo decoders, MIND outperforms the static benchmarks by a large margin and shows minimal performance gap when compared to the neural (Convolutional or Turbo) decoders designed for that particular channel. In addition, MIND also shows strong learning capability for channels not exposed during the meta training phase. …

Hierarchical Reinforcement Learning Algorithm via Multi-Goals Abstraction (HRL-MG)
The recommender system is an important form of intelligent application, which assists users to alleviate from information redundancy. Among the metrics used to evaluate a recommender system, the metric of conversion has become more and more important. The majority of existing recommender systems perform poorly on the metric of conversion due to its extremely sparse feedback signal. To tackle this challenge, we propose a deep hierarchical reinforcement learning based recommendation framework, which consists of two components, i.e., high-level agent and low-level agent. The high-level agent catches long-term sparse conversion signals, and automatically sets abstract goals for low-level agent, while the low-level agent follows the abstract goals and interacts with real-time environment. To solve the inherent problem in hierarchical reinforcement learning, we propose a novel deep hierarchical reinforcement learning algorithm via multi-goals abstraction (HRL-MG). Our proposed algorithm contains three characteristics: 1) the high-level agent generates multiple goals to guide the low-level agent in different stages, which reduces the difficulty of approaching high-level goals; 2) different goals share the same state encoder parameters, which increases the update frequency of the high-level agent and thus accelerates the convergence of our proposed algorithm; 3) an appreciate benefit assignment function is designed to allocate rewards in each goal so as to coordinate different goals in a consistent direction. We evaluate our proposed algorithm based on a real-world e-commerce dataset and validate its effectiveness. …

Heterogeneous Simultaneous Multiscale Change Point Estimator (H-SMUCE)
We propose, a heterogeneous simultaneous multiscale change point estimator called ‘H-SMUCE’ for the detection of multiple change points of the signal in a heterogeneous Gaussian regression model. A piecewise constant function is estimated by minimizing the number of change points over the acceptance region of a multiscale test which locally adapts to changes in the variance. The multiscale test is a combination of local likelihood ratio tests which are properly calibrated by scale-dependent critical values to keep a global nominal level a, even for finite samples. We show that H-SMUCE controls the error of overestimation and underestimation of the number of change points. For this, new deviation bounds for F-type statistics are derived. Moreover, we obtain confidence sets for the whole signal. All results are non-asymptotic and uniform over a large class of heterogeneous change point models. H-SMUCE is fast to compute, achieves the optimal detection rate and estimates the number of change points at almost optimal accuracy for vanishing signals, while still being robust. We compare H-SMUCE with several state of the art methods in simulations and analyse current recordings of a transmembrane protein in the bacterial outer membrane with pronounced heterogeneity for its states. An R-package is available on line. …

Shogun
Shogun is and open-source machine learning library that offers a wide range of efficient and unified machine learning methods. …

Route Constrained Optimization (RCO)
Distillation-based learning boosts the performance of the miniaturized neural network based on the hypothesis that the representation of a teacher model can be used as structured and relatively weak supervision, and thus would be easily learned by a miniaturized model. However, we find that the representation of a converged heavy model is still a strong constraint for training a small student model, which leads to a high lower bound of congruence loss. In this work, inspired by curriculum learning we consider the knowledge distillation from the perspective of curriculum learning by routing. Instead of supervising the student model with a converged teacher model, we supervised it with some anchor points selected from the route in parameter space that the teacher model passed by, as we called route constrained optimization (RCO). We experimentally demonstrate this simple operation greatly reduces the lower bound of congruence loss for knowledge distillation, hint and mimicking learning. On close-set classification tasks like CIFAR100 and ImageNet, RCO improves knowledge distillation by 2.14% and 1.5% respectively. For the sake of evaluating the generalization, we also test RCO on the open-set face recognition task MegaFace. …

Attentional Encoder Network (AEN)
Targeted sentiment classification aims at determining the sentimental tendency towards specific targets. Most of the previous approaches model context and target words using recurrent neural networks such as LSTM in conjunction with attention mechanisms. However, LSTM networks are difficult to parallelize because of their sequential nature. Moreover, since full backpropagation over the sequence requires large amounts of memory, essentially every implementation of backpropagation through time is the truncated version, which brings difficulty in remembering long-term patterns. To address these issues, this paper propose an Attentional Encoder Network (AEN) for targeted sentiment classification. Contrary to previous LSTM based works, AEN eschews complex recurrent neural networks and employs attention based encoders for the modeling between context and target, which can excavate the rich introspective and interactive semantic information from the word embeddings without considering the distance between words. This paper also raise the label unreliability issue and introduce label smoothing regularization term to the loss function for encouraging the model to be less confident with the training labels. Experimental results on three benchmark datasets demonstrate that our model achieves comparable or superior performances with a lightweight model size. …

Scattering Transforms (ScatterNets)
Scattering Transforms (or ScatterNets) introduced by Mallat are a promising start into creating a well-defined feature extractor to use for pattern recognition and image classification tasks. They are of particular interest due to their architectural similarity to Convolutional Neural Networks (CNNs), while requiring no parameter learning and still performing very well (particularly in constrained classification tasks). In this paper we visualize what the deeper layers of a ScatterNet are sensitive to using a ‘DeScatterNet’. We show that the higher orders of ScatterNets are sensitive to complex, edge-like patterns (checker-boards and rippled edges). These complex patterns may be useful for texture classification, but are quite dissimilar from the patterns visualized in second and third layers of Convolutional Neural Networks (CNNs) – the current state of the art Image Classifiers. We propose that this may be the source of the current gaps in performance between ScatterNets and CNNs (83% vs 93% on CIFAR-10 for ScatterNet+SVM vs ResNet). We then use these visualization tools to propose possible enhancements to the ScatterNet design, which show they have the power to extract features more closely resembling CNNs, while still being well-defined and having the invariance properties fundamental to ScatterNets. …

Convolutional Cluster Pooling
We present a novel and hierarchical approach for supervised classification of signals spanning over a fixed graph, reflecting shared properties of the dataset. To this end, we introduce a Convolutional Cluster Pooling layer exploiting a multi-scale clustering in order to highlight, at different resolutions, locally connected regions on the input graph. Our proposal generalises well-established neural models such as Convolutional Neural Networks (CNNs) on irregular and complex domains, by means of the exploitation of the weight sharing property in a graph-oriented architecture. In this work, such property is based on the centrality of each vertex within its soft-assigned cluster. Extensive experiments on NTU RGB+D, CIFAR-10 and 20NEWS demonstrate the effectiveness of the proposed technique in capturing both local and global patterns in graph-structured data out of different domains. …

TensorNetwork
TensorNetwork is an open source library for implementing tensor network algorithms. Tensor networks are sparse data structures originally designed for simulating quantum many-body physics, but are currently also applied in a number of other research areas, including machine learning. We demonstrate the use of the API with applications both physics and machine learning, with details appearing in companion papers.
Introducing TensorNetwork, an Open Source Library for Efficient Tensor Calculations …

Fuzzy bag-of-Words (FBoW)
Recent literature suggests that averaged word vectors followed by simple post-processing outperform many deep learning methods on semantic textual similarity tasks. Furthermore, when averaged word vectors are trained supervised on large corpora of paraphrases, they achieve state-of-the-art results on standard STS benchmarks. Inspired by these insights, we push the limits of word embeddings even further. We propose a novel fuzzy bag-of-words (FBoW) representation for text that contains all the words in the vocabulary simultaneously but with different degrees of membership, which are derived from similarities between word vectors. We show that max-pooled word vectors are only a special case of fuzzy BoW and should be compared via fuzzy Jaccard index rather than cosine similarity. Finally, we propose DynaMax, a completely unsupervised and non-parametric similarity measure that dynamically extracts and max-pools good features depending on the sentence pair. This method is both efficient and easy to implement, yet outperforms current baselines on STS tasks by a large margin and is even competitive with supervised word vectors trained to directly optimise cosine similarity. …

Online Soft Mining (OSM)
Deep metric learning aims to learn a deep embedding that can capture the semantic similarity of data points. Given the availability of massive training samples, deep metric learning is known to suffer from slow convergence due to a large fraction of trivial samples. Therefore, most existing methods generally resort to sample mining strategies for selecting nontrivial samples to accelerate convergence and improve performance. In this work, we identify two critical limitations of the sample mining methods, and provide solutions for both of them. First, previous mining methods assign one binary score to each sample, i.e., dropping or keeping it, so they only selects a subset of relevant samples in a mini-batch. Therefore, we propose a novel sample mining method, called Online Soft Mining (OSM), which assigns one continuous score to each sample to make use of all samples in the mini-batch. OSM learns extended manifolds that preserve useful intraclass variances by focusing on more similar positives. Second, the existing methods are easily influenced by outliers as they are generally included in the mined subset. To address this, we introduce Class-Aware Attention (CAA) that assigns little attention to abnormal data samples. Furthermore, by combining OSM and CAA, we propose a novel weighted contrastive loss to learn discriminative embeddings. Extensive experiments on two fine-grained visual categorisation datasets and two video-based person re-identification benchmarks show that our method significantly outperforms the state-of-the-art. …

Seq2Seq
Multi-label text classification (MLTC) aims to assign multiple labels to each sample in the dataset. The labels usually have internal correlations. However, traditional methods tend to ignore the correlations between labels. In order to capture the correlations between labels, the sequence-to-sequence (Seq2Seq) model views the MLTC task as a sequence generation problem, which achieves excellent performance on this task. However, the Seq2Seq model is not suitable for the MLTC task in essence. The reason is that it requires humans to predefine the order of the output labels, while some of the output labels in the MLTC task are essentially an unordered set rather than an ordered sequence. This conflicts with the strict requirement of the Seq2Seq model for the label order. In this paper, we propose a novel sequence-to-set framework utilizing deep reinforcement learning, which not only captures the correlations between labels, but also reduces the dependence on the label order. Extensive experimental results show that our proposed method outperforms the competitive baselines by a large margin. …

Edge Source Coding
Data compression is an efficient technique to save data storage and transmission costs. However, traditional data compression methods always ignore the impact of user preferences on the statistical distributions of symbols transmitted over the links. Notice that the development of big data technologies and popularization of smart devices enable analyses on user preferences based on data collected from personal handsets. This paper presents a user preference aware lossless data compression method, termed edge source coding, to compress data at the network edge. An optimization problem is formulated to minimize the expected number of bits needed to represent a requested content item in edge source coding. For edge source coding under discrete user preferences, DCA (difference of convex functions programming algorithm) based and k-means++ based algorithms are proposed to give codebook designs. For edge source coding under continuous user preferences, a sampling method is applied to give codebook designs. In addition, edge source coding is extended to the two-user case and codebooks are elaborately designed to utilize multicasting opportunities. Both theoretical analysis and simulations demonstrate the optimal codebook design should take into account user preferences. …

Tensor-Train RNN (TT-RNN)
We present Tensor-Train RNN (TT-RNN), a novel family of neural sequence architectures for multivariate forecasting in environments with nonlinear dynamics. Long-term forecasting in such systems is highly challenging, since there exist long-term temporal dependencies, higher-order correlations and sensitivity to error propagation. Our proposed tensor recurrent architecture addresses these issues by learning the nonlinear dynamics directly using higher order moments and high-order state transition functions. Furthermore, we decompose the higher-order structure using the tensor-train (TT) decomposition to reduce the number of parameters while preserving the model performance. We theoretically establish the approximation properties of Tensor-Train RNNs for general sequence inputs, and such guarantees are not available for usual RNNs. We also demonstrate significant long-term prediction improvements over general RNN and LSTM architectures on a range of simulated environments with nonlinear dynamics, as well on real-world climate and traffic data. …

Deep Frame Interpolation
This work presents a supervised learning based approach to the computer vision problem of frame interpolation. The presented technique could also be used in the cartoon animations since drawing each individual frame consumes a noticeable amount of time. The most existing solutions to this problem use unsupervised methods and focus only on real life videos with already high frame rate. However, the experiments show that such methods do not work as well when the frame rate becomes low and object displacements between frames becomes large. This is due to the fact that interpolation of the large displacement motion requires knowledge of the motion structure thus the simple techniques such as frame averaging start to fail. In this work the deep convolutional neural network is used to solve the frame interpolation problem. In addition, it is shown that incorporating the prior information such as optical flow improves the interpolation quality significantly. …

Deep Distribution Regression
Due to their flexibility and predictive performance, machine-learning based regression methods have become an important tool for predictive modeling and forecasting. However, most methods focus on estimating the conditional mean or specific quantiles of the target quantity and do not provide the full conditional distribution, which contains uncertainty information that might be crucial for decision making. In this article, we provide a general solution by transforming a conditional distribution estimation problem into a constrained multi-class classification problem, in which tools such as deep neural networks. We propose a novel joint binary cross-entropy loss function to accomplish this goal. We demonstrate its performance in various simulation studies comparing to state-of-the-art competing methods. Additionally, our method shows improved accuracy in a probabilistic solar energy forecasting problem. …

Typed Graph Network
Recently, the deep learning community has given growing attention to neural architectures engineered to learn problems in relational domains. Convolutional Neural Networks employ parameter sharing over the image domain, tying the weights of neural connections on a grid topology and thus enforcing the learning of a number of convolutional kernels. By instantiating trainable neural modules and assembling them in varied configurations (apart from grids), one can enforce parameter sharing over graphs, yielding models which can effectively be fed with relational data. In this context, vertices in a graph can be projected into a hyperdimensional real space and iteratively refined over many message-passing iterations in an end-to-end differentiable architecture. Architectures of this family have been referred to with several definitions in the literature, such as Graph Neural Networks, Message-passing Neural Networks, Relational Networks and Graph Networks. In this paper, we revisit the original Graph Neural Network model and show that it generalises many of the recent models, which in turn benefit from the insight of thinking about vertex \textbf{types}. To illustrate the generality of the original model, we present a Graph Neural Network formalisation, which partitions the vertices of a graph into a number of types. Each type represents an entity in the ontology of the problem one wants to learn. This allows – for instance – one to assign embeddings to edges, hyperedges, and any number of global attributes of the graph. As a companion to this paper we provide a Python/Tensorflow library to facilitate the development of such architectures, with which we instantiate the formalisation to reproduce a number of models proposed in the current literature. …

Fenchel Lifted Network
Despite the recent successes of deep neural networks, the corresponding training problem remains highly non-convex and difficult to optimize. Classes of models have been proposed that introduce greater structure to the objective function at the cost of lifting the dimension of the problem. However, these lifted methods sometimes perform poorly compared to traditional neural networks. In this paper, we introduce a new class of lifted models, Fenchel lifted networks, that enjoy the same benefits as previous lifted models, without suffering a degradation in performance over classical networks. Our model represents activation functions as equivalent biconvex constraints and uses Lagrange Multipliers to arrive at a rigorous lower bound of the traditional neural network training problem. This model is efficiently trained using block-coordinate descent and is parallelizable across data points and/or layers. We compare our model against standard fully connected and convolutional networks and show that we are able to match or beat their performance. …

Deep Complex U-Net
Most deep learning-based models for speech enhancement have mainly focused on estimating the magnitude of spectrogram while reusing the phase from noisy speech for reconstruction. This is due to the difficulty of estimating the phase of clean speech. To improve speech enhancement performance, we tackle the phase estimation problem in three ways. First, we propose Deep Complex U-Net, an advanced U-Net structured model incorporating well-defined complex-valued building blocks to deal with complex-valued spectrograms. Second, we propose a polar coordinate-wise complex-valued masking method to reflect the distribution of complex ideal ratio masks. Third, we define a novel loss function, weighted source-to-distortion ratio (wSDR) loss, which is designed to directly correlate with a quantitative evaluation measure. Our model was evaluated on a mixture of the Voice Bank corpus and DEMAND database, which has been widely used by many deep learning models for speech enhancement. Ablation experiments were conducted on the mixed dataset showing that all three proposed approaches are empirically valid. Experimental results show that the proposed method achieves state-of-the-art performance in all metrics, outperforming previous approaches by a large margin. …

Random Labeled Point Process (RLPP)
Missing values frequently arise in modern biomedical studies due to various reasons, including missing tests or complex profiling technologies for different omics measurements. Missing values can complicate the application of clustering algorithms, whose goals are to group points based on some similarity criterion. A common practice for dealing with missing values in the context of clustering is to first impute the missing values, and then apply the clustering algorithm on the completed data. We consider missing values in the context of optimal clustering, which finds an optimal clustering operator with reference to an underlying random labeled point process (RLPP). We show how the missing-value problem fits neatly into the overall framework of optimal clustering by incorporating the missing value mechanism into the random labeled point process and then marginalizing out the missing-value process. In particular, we demonstrate the proposed framework for the Gaussian model with arbitrary covariance structures. Comprehensive experimental studies on both synthetic and real-world RNA-seq data show the superior performance of the proposed optimal clustering with missing values when compared to various clustering approaches. Optimal clustering with missing values obviates the need for imputation-based pre-processing of the data, while at the same time possessing smaller clustering errors. …

KTBoost
In this article, we introduce a novel boosting algorithm called `KTBoost’, which combines kernel boosting and tree boosting. In each boosting iteration, the algorithm adds either a regression tree or reproducing kernel Hilbert space (RKHS) regression function to the ensemble of base learners. Intuitively, the idea is that discontinuous trees and continuous RKHS regression functions complement each other, and that this combination allows for better learning of both continuous and discontinuous functions as well as functions that exhibit parts with varying degrees of regularity. We empirically show that KTBoost outperforms both tree and kernel boosting in terms of predictive accuracy on a wide array of data sets. …

Swapout
We describe Swapout, a new stochastic training method, that outperforms ResNets of identical network structure yielding impressive results on CIFAR-10 and CIFAR-100. Swapout samples from a rich set of architectures including dropout, stochastic depth and residual architectures as special cases. When viewed as a regularization method swapout not only inhibits co-adaptation of units in a layer, similar to dropout, but also across network layers. We conjecture that swapout achieves strong regularization by implicitly tying the parameters across layers. When viewed as an ensemble training method, it samples a much richer set of architectures than existing methods such as dropout or stochastic depth. We propose a parameterization that reveals connections to exiting architectures and suggests a much richer set of architectures to be explored. We show that our formulation suggests an efficient training method and validate our conclusions on CIFAR-10 and CIFAR-100 matching state of the art accuracy. Remarkably, our 32 layer wider model performs similar to a 1001 layer ResNet model. …

Data Illustrator
Data Illustrator: Augmenting Vector Design Tools with Lazy Data Binding for Expressive Visualization Authoring. Building graphical user interfaces for visualization authoring is challenging as one must reconcile the tension between flexible graphics manipulation and procedural visualization generation based on a graphical grammar or declarative languages. To better support designers’ workflows and practices, we propose Data Illustrator, a novel visualization framework. In our approach, all visualizations are initially vector graphics; data binding is applied when necessary and only constrains interactive manipulation to that data bound property. The framework augments graphic design tools with new concepts and operators, and describes the structure and generation of a variety of visualizations. Based on the framework, we design and implement a visualization authoring system. The system extends interaction techniques in modern vector design tools for direct manipulation of visualization configurations and parameters. We demonstrate the expressive power of our approach through a variety of examples. A qualitative study shows that designers can use our framework to compose visualizations. …

First Story Detection (FSD)
Given a series of documents, first story is defined as the first document to discuss a specific event, which occurred at a particular time and place. First story detection (FSD) was firstly defined byAllan in 2002 in terms of topic detection and tracking.
➘ “Novelty Detection”
➘ “Topic Detection and Tracking”
http://…/DePaper.pdf
http://…/storm-first-story-detection …