Variants of Graph Neural Networks (GNNs) for representation learning have been proposed recently and achieved fruitful results in various fields. Among them, graph attention networks (GATs) first employ a self-attention strategy to learn attention weights for each edge in the spatial domain. However, learning the attentions over edges only pays attention to the local information of graphs and greatly increases the number of parameters. In this paper, we first introduce attentions in the spectral domain of graphs. Accordingly, we present Spectral Graph Attention Network (SpGAT) that learn representations for different frequency components regarding weighted filters and graph wavelets bases. In this way, SpGAT can better capture global patterns of graphs in an efficient manner with much fewer learned parameters than that of GAT. We thoroughly evaluate the performance of SpGAT in the semi-supervised node classification task and verified the effectiveness of the learned attentions in the spectral domain.

Throughout the last years, machine learning techniques have been broadly encouraged in the context of deep learning architectures. An interesting algorithm denoted as Restricted Boltzmann Machine relies on energy- and probabilistic-based nature to tackle with the most diverse applications, such as classification, reconstruction, and generation of images and signals. Nevertheless, one can see they are not adequately renowned when compared to other well-known deep learning techniques, e.g., Convolutional Neural Networks. Such behavior promotes the lack of researches and implementations around the literature, coping with the challenge of sufficiently comprehending these energy-based systems. Therefore, in this paper, we propose a Python-inspired framework in the context of energy-based architectures, denoted as Learnergy. Essentially, Learnergy is built upon PyTorch for providing a more friendly environment and a faster prototyping workspace, as well as, possibility the usage of CUDA computations, speeding up their computational time.

Sensor-based human activity recognition has become a critical component of many emerging applications ranging from behavioral medicine to gaming. However, an unprecedented increase in the diversity of sensor devices in the Internet-of-Things era has limited the adoption of activity recognition models for use across different domains. We propose ActiLabel a combinatorial framework that learns structural similarities among the events in an arbitrary domain and those of a different domain. The structural similarities are captured through a graph model, referred to as the it dependency graph, which abstracts details of activity patterns in low-level signal and feature space. The activity labels are then autonomously learned by finding an optimal tiered mapping between the dependency graphs. Extensive experiments based on three public datasets demonstrate the superiority of ActiLabel over state-of-the-art transfer learning and deep learning methods.

Deep learning-based recommendation models are used pervasively and broadly, for example, to recommend movies, products, or other information most relevant to users, in order to enhance the user experience. Among various application domains which have received significant industry and academia research attention, such as image classification, object detection, language and speech translation, the performance of deep learning-based recommendation models is less well explored, even though recommendation tasks unarguably represent significant AI inference cycles at large-scale datacenter fleets. To advance the state of understanding and enable machine learning system development and optimization for the commerce domain, we aim to define an industry-relevant recommendation benchmark for the MLPerf Training andInference Suites. The paper synthesizes the desirable modeling strategies for personalized recommendation systems. We lay out desirable characteristics of recommendation model architectures and data sets. We then summarize the discussions and advice from the MLPerf Recommendation Advisory Board.

This paper studies the problem of post-hoc calibration of machine learning classifiers. We introduce the following desiderata for uncertainty calibration: (a) accuracy-preserving, (b) data-efficient, and (c) high expressive power. We show that none of the existing methods satisfy all three requirements, and demonstrate how Mix-n-Match calibration strategies (i.e., ensemble and composition) can help achieve remarkably better data-efficiency and expressive power while provably preserving classification accuracy of the original classifier. We also show that existing calibration error estimators (e.g., histogram-based ECE) are unreliable especially in small-data regime. Therefore, we propose an alternative data-efficient kernel density-based estimator for a reliable evaluation of the calibration performance and prove its asymptotically unbiasedness and consistency.

Multi-task learning (MTL) aims at improving the generalization performance of several related tasks by leveraging useful information contained in them. However, in industrial scenarios, interpretability is always demanded, and the data of different tasks may be in heterogeneous domains, making the existing methods unsuitable or unsatisfactory. In this paper, following the philosophy of boosted tree, we proposed a two-stage method. In stage one, a common model is built to learn the commonalities using the common features of all instances. Different from the training of conventional boosted tree model, we proposed a regularization strategy and an early-stopping mechanism to optimize the multi-task learning process. In stage two, started by fitting the residual error of the common model, a specific model is constructed with the task-specific instances to further boost the performance. Experiments on both benchmark and real-world datasets validate the effectiveness of the proposed method. What's more, interpretability can be naturally obtained from the tree based method, satisfying the industrial needs.

Reinforcement learning systems require good representations to work well. For decades practical success in reinforcement learning was limited to small domains. Deep reinforcement learning systems, on the other hand, are scalable, not dependent on domain specific prior knowledge and have been successfully used to play Atari, in 3D navigation from pixels, and to control high degree of freedom robots. Unfortunately, the performance of deep reinforcement learning systems is sensitive to hyper-parameter settings and architecture choices. Even well tuned systems exhibit significant instability both within a trial and across experiment replications. In practice, significant expertise and trial and error are usually required to achieve good performance. One potential source of the problem is known as catastrophic interference: when later training decreases performance by overriding previous learning. Interestingly, the powerful generalization that makes Neural Networks (NN) so effective in batch supervised learning might explain the challenges when applying them in reinforcement learning tasks. In this paper, we explore how online NN training and interference interact in reinforcement learning. We find that simply re-mapping the input observations to a high-dimensional space improves learning speed and parameter sensitivity. We also show this preprocessing reduces interference in prediction tasks. More practically, we provide a simple approach to NN training that is easy to implement, and requires little additional computation. We demonstrate that our approach improves performance in both prediction and control with an extensive batch of experiments in classic control domains.

We introduce Deep Adaptive Semantic Logic (DASL), a novel framework for automating the generation of deep neural networks that incorporates user-provided formal knowledge to improve learning from data. We provide formal semantics that demonstrate that our knowledge representation captures all of first order logic and that finite sampling from infinite domains converges to correct truth values. DASL's representation improves on prior neural-symbolic work by avoiding vanishing gradients, allowing deeper logical structure, and enabling richer interactions between the knowledge and learning components. We illustrate DASL through a toy problem in which we add structure to an image classification problem and demonstrate that knowledge of that structure reduces data requirements by a factor of $1000$. We then evaluate DASL on a visual relationship detection task and demonstrate that the addition of commonsense knowledge improves performance by $10.7\%$ in a data scarce setting.

Contextual embeddings, such as ELMo and BERT, move beyond global word representations like Word2Vec and achieve ground-breaking performance on a wide range of natural language processing tasks. Contextual embeddings assign each word a representation based on its context, thereby capturing uses of words across varied contexts and encoding knowledge that transfers across languages. In this survey, we review existing contextual embedding models, cross-lingual polyglot pre-training, the application of contextual embeddings in downstream tasks, model compression, and model analyses.

Multichannel processing is widely used for speech enhancement but several limitations appear when trying to deploy these solutions to the real-world. Distributed sensor arrays that consider several devices with a few microphones is a viable alternative that allows for exploiting the multiple devices equipped with microphones that we are using in our everyday life. In this context, we propose to extend the distributed adaptive node-specific signal estimation approach to a neural networks framework. At each node, a local filtering is performed to send one signal to the other nodes where a mask is estimated by a neural network in order to compute a global multi-channel Wiener filter. In an array of two nodes, we show that this additional signal can be efficiently taken into account to predict the masks and leads to better speech enhancement performances than when the mask estimation relies only on the local signals.