In the last years machine learning (ML) has moved from a academic endeavor to a pervasive technology adopted in almost every aspect of computing. ML-powered products are now embedded in our digital lives: from recommendations of what to watch, to divining our search intent, to powering virtual assistants in consumer and enterprise settings. Recent successes in applying ML in natural sciences revealed that ML can be used to tackle some of the hardest real-world problems humanity faces today. For these reasons ML has become central in the strategy of tech companies and has gathered even more attention from academia than ever before. Despite these successes, what we have witnessed so far is just the beginning. Right now the people training and using ML models are expert developers working within large organizations, but we believe the next wave of ML systems will allow a larger amount of people, potentially without coding skills, to perform the same tasks. These new ML systems will not require users to fully understand all the details of how models are trained and utilized for obtaining predictions. Declarative interfaces are well suited for this goal, by hiding complexity and favouring separation of interests, and can lead to increased productivity. We worked on such abstract interfaces by developing two declarative ML systems, Overton and Ludwig, that require users to declare only their data schema (names and types of inputs) and tasks rather then writing low level ML code. In this article we will describe how ML systems are currently structured, highlight important factors for their success and adoption, what are the issues current ML systems are facing and how the systems we developed addressed them. Finally we will talk about learnings from the development of ML systems throughout the years and how we believe the next generation of ML systems will look like.

Natural Language Processing (NLP) systems have been subjected to a Cambrian explosion of architectural paradigms in the past few years. The scale on the number of contributions and its exponential growth, bring challenges in understanding how NLP architectural patterns evolve and consolidate in different sub-areas and tasks. This paper aims to provide the methodological support for the interpretation of NLP architectural patterns at scale by applying statistical corpus analysis methods over large-scale NLP corpora. We analyse the use of corpus statistics to compute large-scale collocation patterns jointly with graph visualisation methods as a device to interpret architectural patterns at scale. The proposed methods aims to address questions such as: - What is the complete list of architectural patterns present in NLP? - What are the prevailing architectural patterns (classifiers, layers, regularisation, linguistic resources) for each NLP task? - How these patterns are evolving over time and what are the emerging consolidated/canonical architectural motifs?

Future climate change scenarios are usually hypothesized using simulations from weather generators. However, there only a few works comparing and evaluating promising deep learning models for weather generation against classical approaches. This study shows preliminary results making such evaluations for the multisite precipitation synthesis task. We compared two open-source weather generators: IBMWeathergen (an extension of the Weathergen library) and RGeneratePrec, and two deep generative models: GAN and VAE, on a variety of metrics. Our preliminary results can serve as a guide for improving the design of deep learning architectures and algorithms for the multisite precipitation synthesis task.

In this paper, we propose a novel approach to 3D deformable object manipulation leveraging a deep neural network called DeformerNet. Controlling the shape of a 3D object requires an effective state representation that can capture the full 3D geometry of the object. Current methods work around this problem by defining a set of feature points on the object or only deforming the object in 2D image space, which does not truly address the 3D shape control problem. Instead, we explicitly use 3D point clouds as the state representation and apply Convolutional Neural Network on point clouds to learn the 3D features. These features are then mapped to the robot end-effector's position using a fully-connected neural network. Once trained in an end-to-end fashion, DeformerNet directly maps the current point cloud of a deformable object, as well as a target point cloud shape, to the desired displacement in robot gripper position. In addition, we investigate the problem of predicting the manipulation point location given the initial and goal shape of the object.

We introduce a new class of graph neural networks (GNNs), by combining several concepts that were so far studied independently - graph kernels, attention-based networks with structural priors and more recently, efficient Transformers architectures applying small memory footprint implicit attention methods via low rank decomposition techniques. The goal of the paper is twofold. Proposed by us Graph Kernel Attention Transformers (or GKATs) are much more expressive than SOTA GNNs as capable of modeling longer-range dependencies within a single layer. Consequently, they can use more shallow architecture design. Furthermore, GKAT attention layers scale linearly rather than quadratically in the number of nodes of the input graphs, even when those graphs are dense, requiring less compute than their regular graph attention counterparts. They achieve it by applying new classes of graph kernels admitting random feature map decomposition via random walks on graphs. As a byproduct of the introduced techniques, we obtain a new class of learnable graph sketches, called graphots, compactly encoding topological graph properties as well as nodes' features. We conducted exhaustive empirical comparison of our method with nine different GNN classes on tasks ranging from motif detection through social network classification to bioinformatics challenges, showing consistent gains coming from GKATs.

We study the problem of inferring an object-centric scene representation from a single image, aiming to derive a representation that explains the image formation process, captures the scene's 3D nature, and is learned without supervision. Most existing methods on scene decomposition lack one or more of these characteristics, due to the fundamental challenge in integrating the complex 3D-to-2D image formation process into powerful inference schemes like deep networks. In this paper, we propose unsupervised discovery of Object Radiance Fields (uORF), integrating recent progresses in neural 3D scene representations and rendering with deep inference networks for unsupervised 3D scene decomposition. Trained on multi-view RGB images without annotations, uORF learns to decompose complex scenes with diverse, textured background from a single image. We show that uORF performs well on unsupervised 3D scene segmentation, novel view synthesis, and scene editing on three datasets.

We propose a deep learning-based method that uses spatial and temporal information extracted from the sub-6GHz band to predict/track beams in the millimeter-wave (mmWave) band. In more detail, we consider a dual-band communication system operating in both the sub-6GHz and mmWave bands. The objective is to maximize the achievable mutual information in the mmWave band with a hybrid analog/digital architecture where analog precoders (RF precoders) are taken from a finite codebook. Finding a RF precoder using conventional search methods incurs large signalling overhead, and the signalling scales with the number of RF chains and the resolution of the phase shifters. To overcome the issue of large signalling overhead in the mmWave band, the proposed method exploits the spatiotemporal correlation between sub-6GHz and mmWave bands, and it predicts/tracks the RF precoders in the mmWave band from sub-6GHz channel measurements. The proposed method provides a smaller candidate set so that performing a search over that set significantly reduces the signalling overhead compared with conventional search heuristics. Simulations show that the proposed method can provide reasonable achievable rates while significantly reducing the signalling overhead.

As an essential element for the diagnosis and rehabilitation of psychiatric disorders, the electroencephalogram (EEG) based emotion recognition has achieved significant progress due to its high precision and reliability. However, one obstacle to practicality lies in the variability between subjects and sessions. Although several studies have adopted domain adaptation (DA) approaches to tackle this problem, most of them treat multiple EEG data from different subjects and sessions together as a single source domain for transfer, which either fails to satisfy the assumption of domain adaptation that the source has a certain marginal distribution, or increases the difficulty of adaptation. We therefore propose the multi-source marginal distribution adaptation (MS-MDA) for EEG emotion recognition, which takes both domain-invariant and domain-specific features into consideration. First, we assume that different EEG data share the same low-level features, then we construct independent branches for multiple EEG data source domains to adopt one-to-one domain adaptation and extract domain-specific features. Finally, the inference is made by multiple branches. We evaluate our method on SEED and SEED-IV for recognizing three and four emotions, respectively. Experimental results show that the MS-MDA outperforms the comparison methods and state-of-the-art models in cross-session and cross-subject transfer scenarios in our settings. Codes at https://github.com/VoiceBeer/MS-MDA.

Temporal Point Processes (TPPs) are often used to represent the sequence of events ordered as per the time of occurrence. Owing to their flexible nature, TPPs have been used to model different scenarios and have shown applicability in various real-world applications. While TPPs focus on modeling the event occurrence, Marked Temporal Point Process (MTPP) focuses on modeling the category/class of the event as well (termed as the marker). Research in MTPP has garnered substantial attention over the past few years, with an extensive focus on supervised algorithms. Despite the research focus, limited attention has been given to the challenging problem of developing solutions in semi-supervised settings, where algorithms have access to a mix of labeled and unlabeled data. This research proposes a novel algorithm for Semi-supervised Learning for Marked Temporal Point Processes (SSL-MTPP) applicable in such scenarios. The proposed SSL-MTPP algorithm utilizes a combination of labeled and unlabeled data for learning a robust marker prediction model. The proposed algorithm utilizes an RNN-based Encoder-Decoder module for learning effective representations of the time sequence. The efficacy of the proposed algorithm has been demonstrated via multiple protocols on the Retweet dataset, where the proposed SSL-MTPP demonstrates improved performance in comparison to the traditional supervised learning approach.

We introduce Neural Contextual Anomaly Detection (NCAD), a framework for anomaly detection on time series that scales seamlessly from the unsupervised to supervised setting, and is applicable to both univariate and multivariate time series. This is achieved by effectively combining recent developments in representation learning for multivariate time series, with techniques for deep anomaly detection originally developed for computer vision that we tailor to the time series setting. Our window-based approach facilitates learning the boundary between normal and anomalous classes by injecting generic synthetic anomalies into the available data. Moreover, our method can effectively take advantage of all the available information, be it as domain knowledge, or as training labels in the semi-supervised setting. We demonstrate empirically on standard benchmark datasets that our approach obtains a state-of-the-art performance in these settings.