Today, computer systems hold large amounts of personal data. Yet while such an abundance of data allows breakthroughs in AI, and especially machine learning, its existence can be a threat to user privacy, and it can weaken the bonds of trust between humans and AI. Recent regulations now require that, on request, private information about a user must be removed both from computer systems and from machine learning models—this legislation is more colloquially called “the right to be forgotten.” While removing data from back-end databases should be straightforward, it is not sufficient in the AI context as machine learning models often “remember” the old data. Contemporary adversarial attacks on trained models have proven that we can learn whether an instance or an attribute belonged to the training data. This phenomenon calls for a new paradigm, namely machine unlearning , to make machine learning models forget about particular data. It turns out that recent works on machine unlearning have not been able to completely solve the problem due to the lack of common frameworks and resources. Therefore, this article aspires to present a comprehensive examination of machine unlearning’s concepts, designs, methods, and applications. Specifically, as a category collection of cutting-edge studies, the intention behind this article is to serve as a comprehensive resource for researchers and practitioners seeking an introduction to machine unlearning and its formulations, design criteria, removal requests, algorithms, and applications. In addition, we aim to highlight the key findings, current trends, and new research areas that have not yet featured the use of machine unlearning but could benefit greatly from it. We hope that this survey serves as a valuable resource for machine learning researchers and those seeking to innovate privacy technologies. Our resources are publicly available at https://github.com/tamlhp/awesome-machine-unlearning .

With recent advancements in graph neural networks (GNN), GNN-based recommender systems (gRS) have achieved remarkable success in the past few years. Despite this success, existing research reveals that gRSs are still vulnerable to poison attacks , in which the attackers inject fake data to manipulate recommendation results as they desire. This might be due to the fact that existing poison attacks (and countermeasures) are either model-agnostic or specifically designed for traditional recommender algorithms (e.g., neighborhood-based, matrix-factorization-based, or deep-learning-based RSs) that are not gRS. As gRSs are widely adopted in the industry, the problem of how to design poison attacks for gRSs has become a need for robust user experience. Herein, we focus on the use of poison attacks to manipulate item promotion in gRSs. Compared to standard GNNs, attacking gRSs is more challenging due to the heterogeneity of network structure and the entanglement between users and items. To overcome such challenges, we propose GSPAttack —a generative surrogate-based poison attack framework for gRSs. GSPAttack tailors a learning process to surrogate a recommendation model as well as generate fake users and user-item interactions while preserving the data correlation between users and items for recommendation accuracy. Although maintaining high accuracy for other items rather than the target item seems counterintuitive, it is equally crucial to the success of a poison attack. Extensive evaluations on four real-world datasets revealed that GSPAttack outperforms all baselines with competent recommendation performance and is resistant to various countermeasures.

The concentration of fine particulate matter (PM2.5), which represents inhalable particles with diameters of 2.5 micrometers and smaller, is a vital air quality index. Such particles can penetrate deep into the human lungs and severely affect human health. This paper studies accurate PM2.5 prediction, which can potentially contribute to reducing or avoiding the negative consequences. Our approach’s novelty is to utilize the genetic algorithm (GA) and an encoder-decoder (E-D) model for PM2.5 prediction. The GA benefits feature selection and remove outliers to enhance the prediction accuracy. The encoder-decoder model with long short-term memory (LSTM), which relaxes the restrictions between the input and output of the model, can be used to effectively predict the PM2.5 concentration. We evaluate the proposed model on air quality datasets from Hanoi and Taiwan. The evaluation results show that our model achieves excellent performance. By merely using the E-D model, we can obtain more accurate (up to 53.7%) predictions than those of previous works. Moreover, the GA in our model has the advantage of obtaining the optimal feature combination for predicting the PM2.5 concentration. By combining the GA-based feature selection algorithm and the E-D model, our proposed approach further improves the accuracy by at least 13.7%.

Forecasting discharge (Q) and water level (H) are essential factors in hydrological research and flood prediction. In recent years, deep learning has emerged as a viable technique for capturing the non-linear relationship of historical data to generate highly accurate prediction results. Despite the success in various domains, applying deep learning in Q and H prediction is hampered by three critical issues: a shortage of training data, the occurrence of noise in the collected data, and the difficulty in adjusting the model’s hyper-parameters. This work proposes a novel deep learning-based Q–H prediction model that overcomes all the shortcomings encountered by existing approaches. Specifically, to address data scarcity and increase prediction accuracy, we design an ensemble learning architecture that takes advantage of multiple deep learning techniques. Furthermore, we leverage the Singular-Spectrum Analysis (SSA) to remove noise and outliers from the original data. Besides, we exploit the Genetic Algorithm (GA) to propose a novel mechanism that can automatically determine the prediction model’s optimal hyper-parameters. We conducted extensive experiments on two datasets collected from Vietnam’s Red and Dakbla rivers. The results show that our proposed solution outperforms current techniques across a wide range of metrics, including NSE, MSE, MAE, and MAPE. Specifically, by exploiting the ensemble learning technique, we can improve the NSE by at least $$2\%$$2 % . Moreover, with the aid of the SSA-based data preprocessing technique, the NSE is further enhanced by more than $$5\%$$5 % . Finally, thanks to GA-based optimization, our proposed model increases the NSE by at least $$6\%$$6 % and up to $$40\%$$40 % in the best case.

Advances in deep neural network (DNN) architectures have enabled new prediction techniques for stock market data. Unlike other multivariate time-series data, stock markets show two unique characteristics: (i) multi-order dynamics, as stock prices are affected by strong non-pairwise correlations (e.g., within the same industry); and (ii) internal dynamics, as each individual stock shows some particular behaviour. Recent DNN-based methods capture multi-order dynamics using hypergraphs, but rely on the Fourier basis in the convolution, which is both inefficient and ineffective. In addition, they largely ignore internal dynamics by adopting the same model for each stock, which implies a severe information loss. In this paper, we propose a framework for stock movement prediction to overcome the above issues. Specifically, the framework includes temporal generative filters that implement a memory-based mechanism onto an LSTM network in an attempt to learn individual patterns per stock. Moreover, we employ hypergraph attentions to capture the non-pairwise correlations. Here, using the wavelet basis instead of the Fourier basis, enables us to simplify the message passing and focus on the localized convolution. Experiments with US market data over six years show that our framework outperforms state-of-the-art methods in terms of profit and stability. Our source code and data are available at https://github.com/thanhtrunghuynh93/estimate.