20 November 2023 in IEEE Transactions on Knowledge and Data Engineering
With the rising awareness of data assets, data governance, which is to understand where data comes from, how it is collected, and how it is used, has been assuming evergrowing importance. One critical component of data governance gaining increasing attention is auditing machine learning models to determine if specific data has been used for training. Existing auditing techniques, like shadow auditing methods, have shown feasibility under specific conditions such as having access to label information and knowledge of training protocols. However, these conditions are often not met in most real-world applications. In this paper, we introduce a practical framework for auditing data provenance based on a differential mechanism, i.e., after carefully designed transformation, perturbed input data from the target model's training set would result in much more drastic changes in the output than those from the model's non-training set. Our framework is data-dependent and does not require distinguishing training data from non-training data or training additional shadow models with labeled output data. Furthermore, our framework extends beyond point-based data auditing to group-based data auditing, aligning with the needs of realworld applications. Our theoretical analysis of the differential mechanism and the experimental results on real-world data sets verify the proposal's effectiveness. The codes have been uploaded in an anonymous link.
24 April 2025 in IEEE Computational Intelligence Magazine
Symbolic Regression (SR) is a powerful technique for uncovering hidden mathematical expressions from observed data and has broad applications in scientific discovery and automatic programming. Genetic Programming (GP) has traditionally been the dominant technique for solving the SR, benefiting from a robust global search capability that enables the discovery of solutions with high fitting accuracy. Whereas, GP suffers from low search efficiency and may not fully exploit the accumulated knowledge to accelerate convergence. Conversely, deep learning-based methods, particularly those employing Transformer backbones, are trained offline on large-scale datasets. These methods exhibit strong generalization capabilities for unseen tasks without additional training. However, the lack of refinement mechanisms for specific tasks renders them inferior to GP methods in accuracy. This study aims to combine the specific problem-solving capabilities of GP with the generalization strengths of pretrained Transformer models. Specifically, we propose a pretrained model guided GP (PGGP) method, which is a GP-based method that incorporates a pretrained Transformer model to enhance SR problem-solving. New initialization and mutation operators are proposed based on the well-structured equation obtained using the pre-trained model. Extensive experiments are conducted, and the results show that our method not only surpasses comparative methods in terms of accuracy but also reduces the complexity of the generated solutions, potentially enhancing interpretability.
26 June 2023 in Proceedings of the AAAI Conference on Artificial Intelligence
Machine learning models have liberated manpower greatly in many real-world tasks, but their predictions are still worse than humans on some specific instances. To improve the performance, it is natural to optimize machine learning models to take decisions for most instances while delivering a few tricky instances to humans, resulting in the problem of Human Assisted Learning (HAL). Previous works mainly formulated HAL as a constrained optimization problem that tries to find a limited subset of instances for human decision such that the sum of model and human errors can be minimized; and employed the greedy algorithms, whose performance, however, may be limited due to the greedy nature. In this paper, we propose a new framework HAL-EMO based on Evolutionary Multi-objective Optimization, which reformulates HAL as a bi-objective optimization problem that minimizes the number of selected instances for human decision and the total errors simultaneously, and employs a Multi-Objective Evolutionary Algorithm (MOEA) to solve it. We implement HAL-EMO using two MOEAs, the popular NSGA-II as well as the theoretically grounded GSEMO. We also propose a specific MOEA, called BSEMO, with biased selection and balanced mutation for HAL-EMO, and prove that for human assisted regression and classification, HAL-EMO using BSEMO can achieve better and same theoretical guarantees than previous greedy algorithms, respectively. Experiments on the tasks of medical diagnosis and content moderation show the superiority of HAL-EMO (with either NSGA-II, GSEMO or BSEMO) over previous algorithms, and that using BSEMO leads to the best performance of HAL-EMO.
14 August 2024 in IEEE Transactions on Artificial Intelligence
Migration has been a universal phenomenon, which brings opportunities as well as challenges for global development. As the number of migrants (e.g., refugees) increases rapidly, a key challenge faced by each country is the problem of migrant resettlement. This problem has attracted scientific research attention, from the perspective of maximizing the employment rate. Previous works mainly formulated migrant resettlement as an approximately submodular optimization problem subject to multiple matroid constraints and employed the greedy algorithm, whose performance, however, may be limited due to its greedy nature. In this paper, we propose a new framework called Migrant Resettlement by Evolutionary Multi-objective Optimization (MR-EMO), which reformulates migrant resettlement as a bi-objective optimization problem that maximizes the expected number of employed migrants and minimizes the number of dispatched migrants simultaneously, and employs a Multi-Objective Evolutionary Algorithm (MOEA) to solve the bi-objective problem. We implement MR-EMO using three MOEAs: the popular Non-dominated Sorting Genetic Algorithm II (NSGA-II), Multi-Objective Evolutionary Algorithm based on Decomposition (MOEA/D) as well as the theoretically grounded Global Simple Evolutionary Multi-objective Optimizer (GSEMO). To further improve the performance of MR-EMO, we propose a specific MOEA, called Global Simple Evolutionary Multi-objective Optimizer using matrix-Swap mutation and Repair mechanism (GSEMO-SR), which has a better ability to search for feasible solutions. We prove that MR-EMO using either GSEMO or GSEMO-SR can achieve better theoretical guarantees than the previous greedy algorithm. Experimental results under the interview and coordination migration models clearly show the superiority of MR-EMO (with either NSGA-II, MOEA/D, GSEMO or GSEMO-SR) over previous algorithms, and that using GSEMO-SR leads to the best performance of MR-EMO.
13 November 2024 in IEEE Transactions on Artificial Intelligence
Planning the movement of the sink to maximize the lifetime in wireless sensor networks is an essential problem. Many existing mobile sink techniques based on mathematical programming or heuristics have demonstrated the feasibility of the task. Nevertheless, the huge computational cost or the over-reliance on human knowledge can result in relatively low performance. To balance the need for high-quality solutions to minimize inference time, we propose a new framework to construct the movement path of the sink automatically. We cast the lifetime maximization problem as an optimization task within a heterogeneous graph, and learn movement policy for the sink by combining graph neural network with deep reinforcement learning. Our approach comprises three key modules: (1) a heterogeneous graph neural network to learn representations of sites and sensors by aggregating features of neighbor nodes and extracting hierarchical graph features. (2) a multi-head attention mechanism that allows the sites to attend to information from sensor nodes, which highly improves the expressive capacity of the learning model. (3) a greedy policy that learns to append the next best site in the solution incrementally. We design twelve types of static and dynamic maps to simulate different wireless sensor networks in the real world, and extensive experiments are conducted to evaluate and analyze our approach. The empirical results show that our approach consistently outperforms the existing methods on all types of maps. Notably, our approach significantly extends the simulation lifetime without sacrificing a large increase in inference time.