Neural Bandit with Arm Group Graph

Contextual bandits aim to identify among a set of arms the optimal one with the highest reward based on their contextual information. Motivated by the fact that the arms usually exhibit group behaviors and the mutual impacts exist among groups, we introduce a new model, Arm Group Graph (AGG), where the nodes represent the groups of arms and the weighted edges formulate the correlations among groups. To leverage the rich information in AGG, we propose a bandit algorithm, AGG-UCB, where the neural networks are designed to estimate rewards, and we propose to utilize graph neural networks (GNN) to learn the representations of arm groups with correlations. To solve the exploitation-exploration dilemma in bandits, we derive a new upper confidence bound (UCB) built on neural networks (exploitation) for exploration. Furthermore, we prove that AGG-UCB can achieve a near-optimal regret bound with over-parameterized neural networks, and provide the convergence analysis of GNN with fully-connected layers which may be of independent interest. In the end, we conduct extensive experiments against state-of-the-art baselines on multiple public data sets, showing the effectiveness of the proposed algorithm.

Comprehensive Fair Meta-learned Recommender System

In recommender systems, one common challenge is the cold-start problem, where interactions are very limited for fresh users in the systems. To address this challenge, recently, many works introduce the meta-optimization idea into the recommendation scenarios, i.e. learning to learn the user preference by only a few past interaction items. The core idea is to learn global shared meta-initialization parameters for all users and rapidly adapt them into local parameters for each user respectively. They aim at deriving general knowledge across preference learning of various users, so as to rapidly adapt to the future new user with the learned prior and a small amount of training data. However, previous works have shown that recommender systems are generally vulnerable to bias and unfairness. Despite the success of meta-learning at improving the recommendation performance with cold-start, the fairness issues are largely overlooked. In this paper, we propose a comprehensive fair meta-learning framework, named CLOVER, for ensuring the fairness of meta-learned recommendation models. We systematically study three kinds of fairness - individual fairness, counterfactual fairness, and group fairness in the recommender systems, and propose to satisfy all three kinds via a multi-task adversarial learning scheme. Our framework offers a generic training paradigm that is applicable to different meta-learned recommender systems. We demonstrate the effectiveness of CLOVER on the representative meta-learned user preference estimator on three real-world data sets. Empirical results show that CLOVER achieves comprehensive fairness without deteriorating the overall cold-start recommendation performance.

DISCO: Comprehensive and Explainable Disinformation Detection

Disinformation refers to false information deliberately spread to influence the general public, and the negative impact of disinformation on society can be observed for numerous issues, such as political agendas and manipulating financial markets. In this paper, we identify prevalent challenges and advances related to automated disinformation detection from multiple aspects, and propose a comprehensive and explainable disinformation detection framework called DISCO. It leverages the heterogeneity of disinformation and addresses the prediction opaqueness. Then we provide a demonstration of DISCO on a real-world fake news detection task with satisfactory detection accuracy and explanation. The demo video and source code of DISCO is now publicly available. We expect that our demo could pave the way for addressing the limitations of identification, comprehension, and explainability as a whole.

Neural Collaborative Filtering Bandits via Meta Learning

Contextual multi-armed bandits provide powerful tools to solve the exploitation-exploration dilemma in decision making, with direct applications in the personalized recommendation. In fact, collaborative effects among users carry the significant potential to improve the recommendation. In this paper, we introduce and study the problem by exploring `Neural Collaborative Filtering Bandits', where the rewards can be non-linear functions and groups are formed dynamically given different specific contents. To solve this problem, inspired by meta-learning, we propose Meta-Ban (meta-bandits), where a meta-learner is designed to represent and rapidly adapt to dynamic groups, along with a UCB-based exploration strategy. Furthermore, we analyze that Meta-Ban can achieve the regret bound of $\mathcal{O}(\sqrt{T \log T})$, improving a multiplicative factor $\sqrt{\log T}$ over state-of-the-art related works. In the end, we conduct extensive experiments showing that Meta-Ban significantly outperforms six strong baselines.

Training Fair Deep Neural Networks by Balancing Influence

Most fair machine learning methods either highly rely on the sensitive information of the training samples or require a large modification on the target models, which hinders their practical application. To address this issue, we propose a two-stage training algorithm named FAIRIF. It minimizes the loss over the reweighted data set (second stage) where the sample weights are computed to balance the model performance across different demographic groups (first stage). FAIRIF can be applied on a wide range of models trained by stochastic gradient descent without changing the model, while only requiring group annotations on a small validation set to compute sample weights. Theoretically, we show that, in the classification setting, three notions of disparity among different groups can be mitigated by training with the weights. Experiments on synthetic data sets demonstrate that FAIRIF yields models with better fairness-utility trade-offs against various types of bias; and on real-world data sets, we show the effectiveness and scalability of FAIRIF. Moreover, as evidenced by the experiments with pretrained models, FAIRIF is able to alleviate the unfairness issue of pretrained models without hurting their performance.

Adaptive Transfer Learning for Plant Phenotyping

Plant phenotyping (Guo et al. 2021; Pieruschka et al. 2019) focuses on studying the diverse traits of plants related to the plants' growth. To be more specific, by accurately measuring the plant's anatomical, ontogenetical, physiological and biochemical properties, it allows identifying the crucial factors of plants' growth in different environments. One commonly used approach is to predict the plant's traits using hyperspectral reflectance (Yendrek et al. 2017; Wang et al. 2021). However, the data distributions of the hyperspectral reflectance data in plant phenotyping might vary in different environments for different plants. That is, it would be computationally expansive to learn the machine learning models separately for one plant in different environments. To solve this problem, we focus on studying the knowledge transferability of modern machine learning models in plant phenotyping. More specifically, this work aims to answer the following questions. (1) How is the performance of conventional machine learning models, e.g., partial least squares regression (PLSR), Gaussian process regression (GPR) and multi-layer perceptron (MLP), affected by the number of annotated samples for plant phenotyping? (2) Whether could the neural network based transfer learning models improve the performance of plant phenotyping? (3) Could the neural network based transfer learning be improved by using infinite-width hidden layers for plant phenotyping?

From Intrinsic to Counterfactual: On the Explainability of Contextualized Recommender Systems

With the prevalence of deep learning based embedding approaches, recommender systems have become a proven and indispensable tool in various information filtering applications. However, many of them remain difficult to diagnose what aspects of the deep models' input drive the final ranking decision, thus, they cannot often be understood by human stakeholders. In this paper, we investigate the dilemma between recommendation and explainability, and show that by utilizing the contextual features (e.g., item reviews from users), we can design a series of explainable recommender systems without sacrificing their performance. In particular, we propose three types of explainable recommendation strategies with gradual change of model transparency: whitebox, graybox, and blackbox. Each strategy explains its ranking decisions via different mechanisms: attention weights, adversarial perturbations, and counterfactual perturbations. We apply these explainable models on five real-world data sets under the contextualized setting where users and items have explicit interactions. The empirical results show that our model achieves highly competitive ranking performance, and generates accurate and effective explanations in terms of numerous quantitative metrics and qualitative visualizations.

Tackling Oversmoothing of GNNs with Contrastive Learning

Graph neural networks (GNNs) integrate the comprehensive relation of graph data and the representation learning capability of neural networks, which is one of the most popular deep learning methods and achieves state-of-the-art performance in many applications, such as natural language processing and computer vision. In real-world scenarios, increasing the depth (i.e., the number of layers) of GNNs is sometimes necessary to capture more latent knowledge of the input data to mitigate the uncertainty caused by missing values. However, involving more complex structures and more parameters will decrease the performance of GNN models. One reason called oversmoothing is recently introduced but the relevant research remains nascent. In general, oversmoothing makes the final representations of nodes indiscriminative, thus deteriorating the node classification and link prediction performance. In this paper, we first survey the current de-oversmoothing methods and propose three major metrics to evaluate a de-oversmoothing method, i.e., constant divergence indicator, easy-to-determine divergence indicator, and model-agnostic strategy. Then, we propose the Topology-guided Graph Contrastive Layer, named TGCL, which is the first de-oversmoothing method maintaining all three mentioned metrics. With the contrastive learning manner, we provide the theoretical analysis of the effectiveness of the proposed TGCL. Last but not least, we design extensive experiments to illustrate the empirical performance of TGCL comparing with state-of-the-art baselines.

Deep Active Learning by Leveraging Training Dynamics

Active learning theories and methods have been extensively studied in classical statistical learning settings. However, deep active learning, i.e., active learning with deep learning models, is usually based on empirical criteria without solid theoretical justification, thus suffering from heavy doubts when some of those fail to provide benefits in applications. In this paper, by exploring the connection between the generalization performance and the training dynamics, we propose a theory-driven deep active learning method (dynamicAL) which selects samples to maximize training dynamics. In particular, we prove that convergence speed of training and the generalization performance is positively correlated under the ultra-wide condition and show that maximizing the training dynamics leads to a better generalization performance. Further on, to scale up to large deep neural networks and data sets, we introduce two relaxations for the subset selection problem and reduce the time complexity from polynomial to constant. Empirical results show that dynamicAL not only outperforms the other baselines consistently but also scales well on large deep learning models. We hope our work inspires more attempts in bridging the theoretical findings of deep networks and practical impacts in deep active learning applications.