Recurring outbreaks of COVID-19 have posed enduring effects on global society, which calls for a predictor of pandemic waves using various data with early availability. Existing prediction models that forecast the first outbreak wave using mobility data may not be applicable to the multiwave prediction, because the evidence in the USA and Japan has shown that mobility patterns across different waves exhibit varying relationships with fluctuations in infection cases. Therefore, to predict the multiwave pandemic, we propose a Social Awareness-Based Graph Neural Network (SAB-GNN) that considers the decay of symptom-related web search frequency to capture the changes in public awareness across multiple waves. SAB-GNN combines GNN and LSTM to model the complex relationships among urban districts, inter-district mobility patterns, web search history, and future COVID-19 infections. We train our model to predict future pandemic outbreaks in the Tokyo area using its mobility and web search data from April 2020 to May 2021 across four pandemic waves collected by _ANONYMOUS_COMPANY_ under strict privacy protection rules. Results show our model outperforms other baselines including ST-GNN and MPNN+LSTM. Though our model is not computationally expensive (only 3 layers and 10 hidden neurons), the proposed model enables public agencies to anticipate and prepare for future pandemic outbreaks.

Assessing the personality of software engineers may help to match individual traits with the characteristics of development activities such as code review and testing, as well as support managers in team composition. However, self-assessment questionnaires are not a practical solution for collecting multiple observations on a large scale. Instead, automatic personality detection, while overcoming these limitations, is based on off-the-shelf solutions trained on non-technical corpora, which might not be readily applicable to technical domains like Software Engineering (SE). In this paper, we first assess the performance of general-purpose personality detection tools when applied to a technical corpus of developers' emails retrieved from the public archives of the Apache Software Foundation. We observe a general low accuracy of predictions and an overall disagreement among the tools. Second, we replicate two previous research studies in SE by replacing the personality detection tool used to infer developers' personalities from pull-request discussions and emails. We observe that the original results are not confirmed, i.e., changing the tool used in the original study leads to diverging conclusions. Our results suggest a need for personality detection tools specially targeted for the software engineering domain.

Most existing human matting algorithms tried to separate pure human-only foreground from the background. In this paper, we propose a Virtual Multi-modality Foreground Matting (VMFM) method to learn human-object interactive foreground (human and objects interacted with him or her) from a raw RGB image. The VMFM method requires no additional inputs, e.g. trimap or known background. We reformulate foreground matting as a self-supervised multi-modality problem: factor each input image into estimated depth map, segmentation mask, and interaction heatmap using three auto-encoders. In order to fully utilize the characteristics of each modality, we first train a dual encoder-to-decoder network to estimate the same alpha matte. Then we introduce a self-supervised method: Complementary Learning(CL) to predict deviation probability map and exchange reliable gradients across modalities without label. We conducted extensive experiments to analyze the effectiveness of each modality and the significance of different components in complementary learning. We demonstrate that our model outperforms the state-of-the-art methods.

Uncertainty quantification is at the core of the reliability and robustness of machine learning. It is well-known that uncertainty consists of two different types, often referred to as aleatoric and epistemic uncertainties. In this paper, we provide a systematic study on the epistemic uncertainty in deep supervised learning. We rigorously distinguish different sources of epistemic uncertainty, including in particular procedural variability (from the training procedure) and data variability (from the training data). We use our framework to explain how deep ensemble enhances prediction by reducing procedural variability. We also propose two approaches to estimate epistemic uncertainty for a well-trained neural network in practice. One uses influence function derived from the theory of neural tangent kernel that bypasses the convexity assumption violated by modern neural networks. Another uses batching that bypasses the time-consuming Gram matrix inversion in the influence function calculation, while expending minimal re-training effort. We discuss how both approaches overcome some difficulties in applying classical statistical methods to the inference on deep learning.

Most algorithms in classical and contemporary machine learning focus on correlation-based dependence between features to drive performance. Although success has been observed in many relevant problems, these algorithms fail when the underlying causality is inconsistent with the assumed relations. We propose a novel model-agnostic loss function called Causal Loss that improves the interventional quality of the prediction using an intervened neural-causal regularizer. In support of our theoretical results, our experimental illustration shows how causal loss bestows a non-causal associative model (like a standard neural net or decision tree) with interventional capabilities.

Machine unlearning, i.e. having a model forget about some of its training data, has become increasingly more important as privacy legislation promotes variants of the right-to-be-forgotten. In the context of deep learning, approaches for machine unlearning are broadly categorized into two classes: exact unlearning methods, where an entity has formally removed the data point's impact on the model by retraining the model from scratch, and approximate unlearning, where an entity approximates the model parameters one would obtain by exact unlearning to save on compute costs. In this paper we first show that the definition that underlies approximate unlearning, which seeks to prove the approximately unlearned model is close to an exactly retrained model, is incorrect because one can obtain the same model using different datasets. Thus one could unlearn without modifying the model at all. We then turn to exact unlearning approaches and ask how to verify their claims of unlearning. Our results show that even for a given training trajectory one cannot formally prove the absence of certain data points used during training. We thus conclude that unlearning is only well-defined at the algorithmic level, where an entity's only possible auditable claim to unlearning is that they used a particular algorithm designed to allow for external scrutiny during an audit.

Recent studies have empirically investigated different methods to train a stochastic classifier by optimising a PAC-Bayesian bound via stochastic gradient descent. Most of these procedures need to replace the misclassification error with a surrogate loss, leading to a mismatch between the optimisation objective and the actual generalisation bound. The present paper proposes a novel training algorithm that optimises the PAC-Bayesian bound, without relying on any surrogate loss. Empirical results show that the bounds obtained with this approach are tighter than those found in the literature.

Recent evidence has shown the existence of a so-called double-descent and even triple-descent behavior for the generalization error of deep-learning models. This important phenomenon commonly appears in implemented neural network architectures, and also seems to emerge in epoch-wise curves during the training process. A recent line of research has highlighted that random matrix tools can be used to obtain precise analytical asymptotics of the generalization (and training) errors of the random feature model. In this contribution, we analyze the whole temporal behavior of the generalization and training errors under gradient flow for the random feature model. We show that in the asymptotic limit of large system size the full time-evolution path of both errors can be calculated analytically. This allows us to observe how the double and triple descents develop over time, if and when early stopping is an option, and also observe time-wise descent structures. Our techniques are based on Cauchy complex integral representations of the errors together with recent random matrix methods based on linear pencils.

Clustering is an unsupervised data mining technique that can be employed to segment customers. The efficient clustering of customers enables banks to design and make offers based on the features of the target customers. The present study uses a real-world financial dataset (Berka, 2000) to cluster bank customers by an encoder-decoder network and the dynamic time warping (DTW) method. The customer features required for clustering are obtained in four ways: Dynamic Time Warping (DTW), Recency Frequency and Monetary (RFM), LSTM encoder-decoder network, and our proposed hybrid method. Once the LSTM model was trained by customer transaction data, a feature vector of each customer was automatically extracted by the encoder.Moreover, the distance between pairs of sequences of transaction amounts was obtained using DTW. Another vector feature was calculated for customers by RFM scoring. In the hybrid method, the feature vectors are combined from the encoder-decoder output, the DTW distance, and the demographic data (e.g., age and gender). Finally, feature vectors were introduced as input to the k-means clustering algorithm, and we compared clustering results with Silhouette and Davies-Bouldin index. As a result, the clusters obtained from the hybrid approach are more accurate and meaningful than those derived from individual clustering techniques. In addition, the type of neural network layers had a substantial effect on the clusters, and high network error does not necessarily worsen clustering performance.

Modern Deep Neural Networks (DNNs) exhibit impressive generalization properties on a variety of tasks without explicit regularization, suggesting the existence of hidden regularization effects. Recent work by Baratin et al. (2021) sheds light on an intriguing implicit regularization effect, showing that some layers are much more aligned with data labels than other layers. This suggests that as the network grows in depth and width, an implicit layer selection phenomenon occurs during training. In this work, we provide the first explanation for this alignment hierarchy. We introduce and empirically validate the Equilibrium Hypothesis which states that the layers that achieve some balance between forward and backward information loss are the ones with the highest alignment to data labels. Our experiments demonstrate an excellent match with the theoretical predictions.