Mingjian Fu

Yongyan Wen, Siyuan Li, Mingjian Fu et al.

In curriculum reinforcement learning (CRL), an agent incrementally accumulates knowledge over a sequence of tasks (i.e., a curriculum), and the learning process is aimed at using the accumulated knowledge to finally solve a challenging target task. While early CRL works focus on sequencing candidate tasks, recent research explores automatic curriculum generation. Among the rich CRL literature, the interpolation-based CRL paradigm is a main body, which automatically generates intermediate tasks by interpolating between the initial task distribution and the target task distribution in task space with meaningful distance metrics (i.e., can measure the task similarity). However, in challenging navigation tasks, the non-Euclidean context (task) space invalidates this assumption. To achieve automatic curriculum generation in complex task, we propose a novel automatic curriculum generation approach based on measurable task representation learning. To better measure the similarity, we propose to transform the task space to a latent space. Through a variational autoencoder structure that encodes the reward and the state transitions, we achieve a latent task representation with a task similarity measurement property, and two close task embeddings correspond to two similar tasks in terms of rewards and state transitions. Based on the learned task representation, we further develop an automatic curriculum generation scheme, which can effectively generate new tasks more and more similar to the target task. We evaluate our method in a variety of challenging navigation tasks, and the experiment results indicate that the proposed approach surpasses state-of-the-art CRL approaches based on interpolation and generative adversarial networks.

Wentao Chen, Jingtang Chen, Mingjian Fu et al.

Deep reinforcement learning (DRL) finds extensive application in autonomous drone navigation within complex, high-risk environments. However, its practical deployment faces a safety-exploration dilemma: soft penalty mechanisms encourage risky trial-and-error, while most constraint-based methods suffer degraded performance under sensor noise and intent uncertainty. We propose Dynamic-TD3, a physically enhanced framework that enforces strict safety constraints while maintaining maneuverability by modeling navigation as a Constrained Markov Decision Process (CMDP). This framework integrates an Adaptive Trajectory Relational Evolution Mechanism (ATREM) to capture long-range intentions and employs a Physically Aware Gated Kalman Filter (PAG-KF) to mitigate non-stationary observation noise. The resulting state representation drives a dual-criterion policy that balances mission efficiency against hard safety constraints via Lagrangian relaxation. In experiments with aggressive dynamic threats, this approach demonstrates superior collision avoidance performance, reduced energy consumption, and smoother flight trajectories.

Mingjian Fu, Hengsheng Chen, Dongchun Jiang et al.

In the era of advancing information technology, recommender systems have emerged as crucial tools for dealing with information overload. However, traditional recommender systems still have limitations in capturing the dynamic evolution of user behavior. To better understand and predict user behavior, especially taking into account the complexity of temporal evolution, sequential recommender systems have gradually become the focus of research. Currently, many sequential recommendation algorithms ignore the amplification effects of prevalent biases, which leads to recommendation results being susceptible to the Matthew Effect. Additionally, it will impose limitations on the recommender system's ability to deeply perceive and capture the dynamic shifts in user preferences, thereby diminishing the extent of its recommendation reach. To address this issue effectively, we propose a recommendation system based on sequential information and attention mechanism called Multi-Perspective Attention Bias Sequential Recommendation (MABSRec). Firstly, we reconstruct user sequences into three short types and utilize graph neural networks for item weighting. Subsequently, an adaptive multi-bias perspective attention module is proposed to enhance the accuracy of recommendations. Experimental results show that the MABSRec model exhibits significant advantages in all evaluation metrics, demonstrating its excellent performance in the sequence recommendation task.

Lanyu Yang, Dongchun Jiang, Fuqiang Guo et al.

Cooperative behavior is prevalent in both human society and nature. Understanding the emergence and maintenance of cooperation among self-interested individuals remains a significant challenge in evolutionary biology and social sciences. Reinforcement learning (RL) provides a suitable framework for studying evolutionary game theory as it can adapt to environmental changes and maximize expected benefits. In this study, we employ the State-Action-Reward-State-Action (SARSA) algorithm as the decision-making mechanism for individuals in evolutionary game theory. Initially, we apply SARSA to imitation learning, where agents select neighbors to imitate based on rewards. This approach allows us to observe behavioral changes in agents without independent decision-making abilities. Subsequently, SARSA is utilized for primary agents to independently choose cooperation or betrayal with their neighbors. We evaluate the impact of SARSA on cooperation rates by analyzing variations in rewards and the distribution of cooperators and defectors within the network.