With the rapid development of computing power network (CPN), resource usage, data exchange, and data aggregation among edge computing nodes have become critical aspects of optimizing resource scheduling. However, such data contains sensitive information, including computational task characteristics and user behavior patterns, making it susceptible to privacy leakage risks. To address this challenge, we propose a privacy-preserving data aggregation (PPDA) scheme that integrates dual secret sharing and an improved threshold Schnorr signature. Our scheme realizes dual masking through Shamir’s Secret Sharing (SSS). It ensures data integrity and source authenticity via threshold signatures. Furthermore, we introduce blockchain smart contracts to enable dynamic updates of secret shares and verifiable publication of aggregation results. In addition to having fault tolerance, the scheme significantly improves the aggregation efficiency in large-scale CPN environments. Theoretical analysis demonstrates that our solution can effectively resist both internal/external attacks and collusion attacks, ensuring the confidentiality of single-node data. Performance evaluations further indicate its computational efficiency surpasses comparable schemes, particularly in large-scale CPN environments.

We introduce an automatic curriculum algorithm, Variational Automatic Curriculum Learning (VACL), for solving challenging goal-conditioned cooperative multi-agent reinforcement learning problems. We motivate our curriculum learning paradigm through a variational perspective, where the learning objective can be decomposed into two terms: task learning on the current curriculum, and curriculum update to a new task distribution. Local optimization over the second term suggests that the curriculum should gradually expand the training tasks from easy to hard. Our VACL algorithm implements this variational paradigm with two practical components, task expansion and entity curriculum, which produces a series of training tasks over both the task configurations as well as the number of entities in the task. Experiment results show that VACL solves a collection of sparse-reward problems with a large number of agents. Particularly, using a single desktop machine, VACL achieves 98% coverage rate with 100 agents in the simple-spread benchmark and reproduces the ramp-use behavior originally shown in OpenAI’s hide-and-seek project.

Multi-agent reinforcement learning (MARL) is of importance for variable real-world applications but remains more challenges like stationarity and scalability. While recently value function factorization methods have obtained empirical good results in cooperative multi-agent environment, these works mostly focus on the decomposable learning structures. Inspired by the application of attention mechanism in machine translation and other related domains, we propose an attention based approach called attention value decomposition network (AVD-Net), which capitalizes on the coordination relations between agents. AVD-Net employs centralized training with decentralized execution (CTDE) paradigm, which factorizes the joint action-value functions with only local observations and actions of agents. Our method is evaluated on multi-agent particle environment (MPE) and StarCraft micromanagement environment (SMAC). The experiment results show the strength of our approach compared to existing methods with state-of-the-art performance in cooperative scenarios.