Graph Neural Networks (GNNs) have been shown to be effective models for different predictive tasks on graph-structured data. Recent work on their expressive power has focused on isomorphism tasks and countable feature spaces. We extend this theoretical framework to include continuous features - which occur regularly in real-world input domains and within the hidden layers of GNNs - and we demonstrate the requirement for multiple aggregation functions in this context. Accordingly, we propose Principal Neighbourhood Aggregation (PNA), a novel architecture combining multiple aggregators with degree-scalers (which generalize the sum aggregator). Finally, we compare the capacity of different models to capture and exploit the graph structure via a novel benchmark containing multiple tasks taken from classical graph theory, alongside existing benchmarks from real-world domains, all of which demonstrate the strength of our model. With this work, we hope to steer some of the GNN research towards new aggregation methods which we believe are essential in the search for powerful and robust models.

Noname manuscript No. (will be inserted by the editor) Abstract Video Question Answering (Video QA) challenges show consistent improvements over state-of-the-art methods modelers in multiple fronts. Modeling video necessitates on well-studied benchmarks including large-scale real-world building not only spatiotemporal models for the dynamic datasets such as TGIF-QA and TVQA, demonstrating the visual channel but also multimodal structures for associated strong capabilities of our CRN unit and the HCRN for complex information channels such as subtitles or audio. To the best of our knowledge, adds at least two more layers of complexity - selecting relevant the HCRN is the very first method attempting to handle content for each channel in the context of the linguistic long and short-form multimodal Video QA at the same time. To address these modules · Hierarchy requirements, we start with two insights: (a) content selection and relation construction can be jointly encapsulated into a conditional computational structure, and (b) video-length 1 Introduction structures can be composed hierarchically. For (a) this paper introduces a general-reusable reusable neural unit dubbed Answering natural questions about a video is a powerful Conditional Relation Network (CRN) taking as input a set of demonstration of cognitive capability. The task involves acquisition tensorial objects and translating into a new set of objects that and manipulation of spatiotemporal visual, acoustic encode relations of the inputs. The generic design of CRN and linguistic representations from the video guided by helps ease the common complex model building process the compositional semantics of linguistic cues [1, 2, 3, 4, 5, of Video QA by simple block stacking and rearrangements 6]. As questions are potentially unconstrained, Video QA with flexibility in accommodating diverse input modalities requires deep modeling capacity to encode and represent crucial and conditioning features across both visual and linguistic multimodal video properties such as linguistic content, domains. As a result, we realize insight (b) by introducing object permanence, motion profiles, prolonged actions, and Hierarchical Conditional Relation Networks (HCRN) for varying-length temporal relations in a hierarchical manner. The HCRN primarily aims at exploiting intrinsic For Video QA, the visual and textual representations should properties of the visual content of a video as well as its accompanying ideally be question-specific and answer-ready.

We present our HABERTOR model for detecting hatespeech in large scale user-generated content. Inspired by the recent success of the BERT model, we propose several modifications to BERT to enhance the performance on the downstream hatespeech classification task. HABERTOR inherits BERT's architecture, but is different in four aspects: (i) it generates its own vocabularies and is pre-trained from the scratch using the largest scale hatespeech dataset; (ii) it consists of Quaternion-based factorized components, resulting in a much smaller number of parameters, faster training and inferencing, as well as less memory usage; (iii) it uses our proposed multi-source ensemble heads with a pooling layer for separate input sources, to further enhance its effectiveness; and (iv) it uses a regularized adversarial training with our proposed fine-grained and adaptive noise magnitude to enhance its robustness. Through experiments on the large-scale real-world hatespeech dataset with 1.4M annotated comments, we show that HABERTOR works better than 15 state-of-the-art hatespeech detection methods, including fine-tuning Language Models. In particular, comparing with BERT, our HABERTOR is 4~5 times faster in the training/inferencing phase, uses less than 1/3 of the memory, and has better performance, even though we pre-train it by using less than 1% of the number of words. Our generalizability analysis shows that HABERTOR transfers well to other unseen hatespeech datasets and is a more efficient and effective alternative to BERT for the hatespeech classification.

There is considerable evidence that deep neural networks are vulnerable to adversarial perturbations applied directly to their digital inputs. However, it remains an open question whether this translates to vulnerabilities in real-world systems. Specifically, in the context of image inputs to autonomous driving systems, an attack can be achieved only by modifying the physical environment, so as to ensure that the resulting stream of video inputs to the car's controller leads to incorrect driving decisions. Inducing this effect on the video inputs indirectly through the environment requires accounting for system dynamics and tracking viewpoint changes. We propose a scalable and efficient approach for finding adversarial physical modifications, using a differentiable approximation for the mapping from environmental modifications-namely, rectangles drawn on the road-to the corresponding video inputs to the controller network. Given the color, location, position, and orientation parameters of the rectangles, our mapping composites them onto pre-recorded video streams of the original environment. Our mapping accounts for geometric and color variations, is differentiable with respect to rectangle parameters, and uses multiple original video streams obtained by varying the driving trajectory. When combined with a neural network-based controller, our approach allows the design of adversarial modifications through end-to-end gradient-based optimization. We evaluate our approach using the Carla autonomous driving simulator, and show that it is significantly more scalable and far more effective at generating attacks than a prior black-box approach based on Bayesian Optimization.

Four years ago, an experimental system known as PilotNet became the first NVIDIA system to steer an autonomous car along a roadway. This system represents a departure from the classical approach for self-driving in which the process is manually decomposed into a series of modules, each performing a different task. In PilotNet, on the other hand, a single deep neural network (DNN) takes pixels as input and produces a desired vehicle trajectory as output; there are no distinct internal modules connected by human-designed interfaces. We believe that handcrafted interfaces ultimately limit performance by restricting information flow through the system and that a learned approach, in combination with other artificial intelligence systems that add redundancy, will lead to better overall performing systems. We continue to conduct research toward that goal. This document describes the PilotNet lane-keeping effort, carried out over the past five years by our NVIDIA PilotNet group in Holmdel, New Jersey. Here we present a snapshot of system status in mid-2020 and highlight some of the work done by the PilotNet group.

Training a deep neural network (DNN) via federated learning allows participants to share model updates (gradients), instead of the data itself. However, recent studies show that unintended latent information (e.g. gender or race) carried by the gradients can be discovered by attackers, compromising the promised privacy guarantee of federated learning. Existing privacy-preserving techniques (e.g. differential privacy) either have limited defensive capacity against the potential attacks, or suffer from considerable model utility loss. Moreover, characterizing the latent information carried by the gradients and the consequent privacy leakage has been a major theoretical and practical challenge. In this paper, we propose two new metrics to address these challenges: the empirical $\mathcal{V}$-information, a theoretically grounded notion of information which measures the amount of gradient information that is usable for an attacker, and the sensitivity analysis that utilizes the Jacobian matrix to measure the amount of changes in the gradients with respect to latent information which further quantifies private risk. We show that these metrics can localize the private information in each layer of a DNN and quantify the leakage depending on how sensitive the gradients are with respect to the latent information. As a practical application, we design LatenTZ: a federated learning framework that lets the most sensitive layers to run in the clients' Trusted Execution Environments (TEE). The implementation evaluation of LatenTZ shows that TEE-based approaches are promising for defending against powerful property inference attacks without a significant overhead in the clients' computing resources nor trading off the model's utility.

Abstract--Point clouds are often sparse and incomplete, which imposes difficulties for real-world applications, such as 3D object classification, detection and segmentation. Existing shape completion methods tend to generate coarse shapes of objects without finegrained details. Moreover, current approaches require fully-complete ground truth, which are difficult to obtain in real-world applications. In view of these, we propose a self-supervised object completion method, which optimizes the training procedure solely on the partial input without utilizing the fully-complete ground truth. In order to generate high-quality objects with detailed geometric structures, we propose a cascaded refinement network (CRN) with a coarse-to-fine strategy to synthesize the complete objects. Considering the local details of partial input together with the adversarial training, we are able to learn the complicated distributions of point clouds and generate the object details as realistic as possible. We verify our self-supervised method on both unsupervised and supervised experimental settings and show superior performances. Quantitative and qualitative experiments on different datasets demonstrate that our method achieves more realistic outputs compared to existing state-of-the-art approaches on the 3D point cloud completion task. A large amount of works [1], [2], [3], [4] have been proposed for point cloud analysis by directly extracting pointwise features from the point coordinates.

In recent years, machine learning algorithms have been applied widely in various fields such as health, transportation, and the autonomous car. With the rapid developments of deep learning techniques, it is critical to take the security concern into account for the application of the algorithms. While machine learning offers significant advantages in terms of the application of algorithms, the issue of security is ignored. Since it has many applications in the real world, security is a vital part of the algorithms. In this paper, we have proposed a mitigation method for adversarial attacks against machine learning models with an autoencoder model that is one of the generative ones. The main idea behind adversarial attacks against machine learning models is to produce erroneous results by manipulating trained models. We have also presented the performance of autoencoder models to various attack methods from deep neural networks to traditional algorithms by using different methods such as non-targeted and targeted attacks to multi-class logistic regression, a fast gradient sign method, a targeted fast gradient sign method and a basic iterative method attack to neural networks for the MNIST dataset.

This paper studies the novel concept of weight correlation in deep neural networks and discusses its impact on the networks' generalisation ability. For fully-connected layers, the weight correlation is defined as the average cosine similarity between weight vectors of neurons, and for convolutional layers, the weight correlation is defined as the cosine similarity between filter matrices. Theoretically, we show that, weight correlation can, and should, be incorporated into the PAC Bayesian framework for the generalisation of neural networks, and the resulting generalisation bound is monotonic with respect to the weight correlation. We formulate a new complexity measure, which lifts the PAC Bayes measure with weight correlation, and experimentally confirm that it is able to rank the generalisation errors of a set of networks more precisely than existing measures. More importantly, we develop a new regulariser for training, and provide extensive experiments that show that the generalisation error can be greatly reduced with our novel approach.

Nowadays, there are many types of neural networks in deep learning which are used for different purposes. In this article, we will go through the most used topologies in neural networks, briefly introduce how they work, along with some of their applications to real-world challenges. This article is our third tutorial on neural networks, to start with our first one, check out neural networks from scratch with Python code and math in detail. The perceptron model is also known as a single-layer neural network. In this type of neural network, there are no hidden layers.