Generative models trained in an unsupervised manner may set high likelihood and low reconstruction loss to Out-of-Distribution (OoD) samples. This increases Type II errors and leads to missed anomalies, overall decreasing Anomaly Detection (AD) performance. In addition, AD models underperform due to the rarity of anomalies. To address these limitations, we propose the OoD Minimum Anomaly Score GAN (OMASGAN). OMASGAN generates, in a negative data augmentation manner, anomalous samples on the estimated distribution boundary. These samples are then used to refine an AD model, leading to more accurate estimation of the underlying data distribution including multimodal supports with disconnected modes. OMASGAN performs retraining by including the abnormal minimum-anomaly-score OoD samples generated on the distribution boundary in a self-supervised learning manner. For inference, for AD, we devise a discriminator which is trained with negative and positive samples either generated (negative or positive) or real (only positive). OMASGAN addresses the rarity of anomalies by generating strong and adversarial OoD samples on the distribution boundary using only normal class data, effectively addressing mode collapse. A key characteristic of our model is that it uses any f-divergence distribution metric in its variational representation, not requiring invertibility. OMASGAN does not use feature engineering and makes no assumptions about the data distribution. The evaluation of OMASGAN on image data using the leave-one-out methodology shows that it achieves an improvement of at least 0.24 and 0.07 points in AUROC on average on the MNIST and CIFAR-10 datasets, respectively, over other benchmark and state-of-the-art models for AD.

Encoder-decoder networks with attention have proven to be a powerful way to solve many sequence-to-sequence tasks. In these networks, attention aligns encoder and decoder states and is often used for visualizing network behavior. However, the mechanisms used by networks to generate appropriate attention matrices are still mysterious. Moreover, how these mechanisms vary depending on the particular architecture used for the encoder and decoder (recurrent, feed-forward, etc.) are also not well understood. In this work, we investigate how encoder-decoder networks solve different sequence-to-sequence tasks. We introduce a way of decomposing hidden states over a sequence into temporal (independent of input) and input-driven (independent of sequence position) components. This reveals how attention matrices are formed: depending on the task requirements, networks rely more heavily on either the temporal or input-driven components. These findings hold across both recurrent and feed-forward architectures despite their differences in forming the temporal components. Overall, our results provide new insight into the inner workings of attention-based encoder-decoder networks.

Modelling interactions is critical in learning complex dynamical systems, namely systems of interacting objects with highly non-linear and time-dependent behaviour. A large class of such systems can be formalized as $\textit{geometric graphs}$, $\textit{i.e.}$, graphs with nodes positioned in the Euclidean space given an $\textit{arbitrarily}$ chosen global coordinate system, for instance vehicles in a traffic scene. Notwithstanding the arbitrary global coordinate system, the governing dynamics of the respective dynamical systems are invariant to rotations and translations, also known as $\textit{Galilean invariance}$. As ignoring these invariances leads to worse generalization, in this work we propose local coordinate frames per node-object to induce roto-translation invariance to the geometric graph of the interacting dynamical system. Further, the local coordinate frames allow for a natural definition of anisotropic filtering in graph neural networks. Experiments in traffic scenes, 3D motion capture, and colliding particles demonstrate that the proposed approach comfortably outperforms the recent state-of-the-art.

Deep Metric Learning (DML) plays a critical role in various machine learning tasks. However, most existing deep metric learning methods with binary similarity are sensitive to noisy labels, which are widely present in real-world data. Since these noisy labels often cause severe performance degradation, it is crucial to enhance the robustness and generalization ability of DML. In this paper, we propose an Adaptive Hierarchical Similarity Metric Learning method. It considers two noise-insensitive information, \textit{i.e.}, class-wise divergence and sample-wise consistency. Specifically, class-wise divergence can effectively excavate richer similarity information beyond binary in modeling by taking advantage of Hyperbolic metric learning, while sample-wise consistency can further improve the generalization ability of the model using contrastive augmentation. More importantly, we design an adaptive strategy to integrate this information in a unified view. It is noteworthy that the new method can be extended to any pair-based metric loss. Extensive experimental results on benchmark datasets demonstrate that our method achieves state-of-the-art performance compared with current deep metric learning approaches.

Consumer Event-Cause Extraction, the task aimed at extracting the potential causes behind certain events in the text, has gained much attention in recent years due to its wide applications. The ICDM 2020 conference sets up an evaluation competition that aims to extract events and the causes of the extracted events with a specified subject (a brand or product). In this task, we mainly focus on how to construct an end-to-end model, and extract multiple event types and event-causes simultaneously. To this end, we introduce a fresh perspective to revisit the relational event-cause extraction task and propose a novel sequence tagging framework, instead of extracting event types and events-causes separately. Experiments show our framework outperforms baseline methods even when its encoder module uses an initialized pre-trained BERT encoder, showing the power of the new tagging framework. In this competition, our team achieved 1st place in the first stage leaderboard, and 3rd place in the final stage leaderboard.

Image anomaly detection is a binary classification problem that decides whether an input image contains an anomaly or not. Image anomaly localization is to further localize the anomalous region at the pixel level. Due to recent advances in deep learning and availability of new datasets, recent research works are no longer limited to the image-level anomaly detection result, but also show a significant interest in the pixel-level localization of anomaly regions. Image anomaly detection and localization find real-world applications such as manufacturing process monitoring[1], medical image analysis [2, 3], and video surveillance analysis [4, 5]. Most well-studied localization solutions (e.g., semantic segmentation) rely on heavy supervision, where a large number of pixel-level labels and many labeled images are needed. However, in the context of image anomaly detection and localization, a typical assumption is that only normal (i.e., artifact-free) images are available in the training stage.

Recent years have witnessed an increasing use of coordinated accounts on social media, operated by misinformation campaigns to influence public opinion and manipulate social outcomes. Consequently, there is an urgent need to develop an effective methodology for coordinated group detection to combat the misinformation on social media. However, existing works suffer from various drawbacks, such as, either limited performance due to extreme reliance on predefined signatures of coordination, or instead an inability to address the natural sparsity of account activities on social media with useful prior domain knowledge. Therefore, in this paper, we propose a coordination detection framework incorporating neural temporal point process with prior knowledge such as temporal logic or pre-defined filtering functions. Specifically, when modeling the observed data from social media with neural temporal point process, we jointly learn a Gibbs-like distribution of group assignment based on how consistent an assignment is to (1) the account embedding space and (2) the prior knowledge. To address the challenge that the distribution is hard to be efficiently computed and sampled from, we design a theoretically guaranteed variational inference approach to learn a mean-field approximation for it. Experimental results on a real-world dataset show the effectiveness of our proposed method compared to the SOTA model in both unsupervised and semi-supervised settings. We further apply our model on a COVID-19 Vaccine Tweets dataset. The detection result suggests the presence of suspicious coordinated efforts on spreading misinformation about COVID-19 vaccines.

In this work, we propose a unified framework, called Visual Reasoning with Differ-entiable Physics (VRDP), that can jointly learn visual concepts and infer physics models of objects and their interactions from videos and language. This is achieved by seamlessly integrating three components: a visual perception module, a concept learner, and a differentiable physics engine. The visual perception module parses each video frame into object-centric trajectories and represents them as latent scene representations. The concept learner grounds visual concepts (e.g., color, shape, and material) from these object-centric representations based on the language, thus providing prior knowledge for the physics engine. The differentiable physics model, implemented as an impulse-based differentiable rigid-body simulator, performs differentiable physical simulation based on the grounded concepts to infer physical properties, such as mass, restitution, and velocity, by fitting the simulated trajectories into the video observations. Consequently, these learned concepts and physical models can explain what we have seen and imagine what is about to happen in future and counterfactual scenarios. Integrating differentiable physics into the dynamic reasoning framework offers several appealing benefits. More accurate dynamics prediction in learned physics models enables state-of-the-art performance on both synthetic and real-world benchmarks while still maintaining high transparency and interpretability; most notably, VRDP improves the accuracy of predictive and counterfactual questions by 4.5% and 11.5% compared to its best counterpart. VRDP is also highly data-efficient: physical parameters can be optimized from very few videos, and even a single video can be sufficient. Finally, with all physical parameters inferred, VRDP can quickly learn new concepts from a few examples.

Modern machine learning models are complicated. Most of them rely on convoluted latent representations of their input to issue a prediction. To achieve greater transparency than a black-box that connects inputs to predictions, it is necessary to gain a deeper understanding of these latent representations. To that aim, we propose SimplEx: a user-centred method that provides example-based explanations with reference to a freely selected set of examples, called the corpus. SimplEx uses the corpus to improve the user's understanding of the latent space with post-hoc explanations answering two questions: (1) Which corpus examples explain the prediction issued for a given test example? (2) What features of these corpus examples are relevant for the model to relate them to the test example? SimplEx provides an answer by reconstructing the test latent representation as a mixture of corpus latent representations. Further, we propose a novel approach, the Integrated Jacobian, that allows SimplEx to make explicit the contribution of each corpus feature in the mixture. Through experiments on tasks ranging from mortality prediction to image classification, we demonstrate that these decompositions are robust and accurate. With illustrative use cases in medicine, we show that SimplEx empowers the user by highlighting relevant patterns in the corpus that explain model representations. Moreover, we demonstrate how the freedom in choosing the corpus allows the user to have personalized explanations in terms of examples that are meaningful for them.

Communication requires having a common language, a lingua franca, between agents. This language could emerge via a consensus process, but it may require many generations of trial and error. Alternatively, the lingua franca can be given by the environment, where agents ground their language in representations of the observed world. We demonstrate a simple way to ground language in learned representations, which facilitates decentralized multi-agent communication and coordination. We find that a standard representation learning algorithm -- autoencoding -- is sufficient for arriving at a grounded common language. When agents broadcast these representations, they learn to understand and respond to each other's utterances and achieve surprisingly strong task performance across a variety of multi-agent communication environments.