Automated planning is a prominent area of Artificial Intelligence, and an important component for intelligent autonomous agents. A critical aspect of domain-independent planning is the domain model, that encodes a formal representation of domain knowledge needed to reason about a given problem. Despite the crucial role of domain models in automated planning, there is a lack of tools supporting the knowledge engineering process by comparing different versions of the models, in particular, determining and highlighting differences the models have.In this paper, we formalise the notions of structural equivalence and strong equivalence of domain models, and, on top of these, we introduce a novel concept of similarity of domain models. To measure the similarity of two models, we introduce a directed graph representation of lifted domain models that allows the formulation of the domain model similarity problem as a variant of the graph edit distance problem. We propose an Answer Set Programming approach to optimally solve the domain model similarity problem, that identifies the minimum number of modifications the models need to become strongly equivalent, and we demonstrate the capabilities of the approach on a range of benchmark models.
Successful human-agent teaming relies on an agent being able to understand instructions given by a (human) principal. In many cases, an instruction may be incomplete or ambiguous. In such cases, the agent must infer the unspoken intentions from their shared context, that is, it must exercise the principal’s Theory of Mind (ToM) and infer the mental states of its principal. We consider the prospects of effective human-agent collaboration using large language models (LLMs). To assess ToM in a dynamic, goal-oriented, and collaborative environment, we introduce a novel task, Instruction Inference, in which an agent assists a principal in reaching a goal by interpreting incomplete or ambiguous instructions.We present Tomcat, an LLM-based agent, designed to exhibit ToM reasoning in interpreting and responding to the principal’s instructions. We implemented two variants of Tomcat. One, dubbed Fs-CoT (Fs for few-shot, CoT for chain-of-thought), is based on a small number of examples demonstrating the requisite structured reasoning. One, dubbed CP (commonsense prompt), relies on commonsense knowledge and information about the problem. We realized both variants of Tomcat on three leading LLMs, namely, GPT-4o, DeepSeek-R1, and Gemma-3-27B. To evaluate the effectiveness of Tomcat, we conducted a study with 52 human participants in which we provided participants with the same information as the CP variant. We computed intent accuracy, action optimality, and planning optimality to measure the ToM capabilities of Tomcat and our study participants. We found that Tomcat with Fs-CoT, particularly with GPT-4o and DeepSeek-R1, achieves performance comparable to the human participants, underscoring its ToM potential for human-agent collaboration.
With the tremendous expansion of graph-structured data, graph neural networks (GNNs) have emerged as the essential tools to tackle graph-related problems. Existing methods generally assume an ideal setting where all class labels are clean. However, due to the label noise and expensive annotation cost, many real-world graph labels are incorrect and scarce. Directly training GNNs under such supervision may lead to overfitting and poor generalization. Despite the effectiveness achieved by existing approaches for handling noisy labels in vision domains, the study on graphs remains largely under-explored. In this paper, we explore a significant yet challenging scenario where nodes on the graph are sparsely and noisily labeled. To address these issues, we propose an evidence-guided framework (EvGNN) to mitigate the negative impact of noisy labels and enhance the utilization of unlabeled nodes on graphs in a multi-expert fashion. Specifically, EvGNN leverages the idea of subjective logic guided by the Dempster-Shafer evidence theory to provide evidence and quantify uncertainty for each expert. Then a novel fusion strategy is developed to overcome highly conflicting phenomena among experts. Moreover, to boost supervisory signals on the graph, we further propose consistency learning and pseudo-labeling as regularizations to fully leverage the unlabeled nodes, thus providing auxiliary supervision against sparse labels. We show the effectiveness of our proposed method through extensive experiments on real-world datasets.
Federated graph convolutional network (FedGCN) is an emerging artificial intelligence technique in recommender systems, which allows each user to hold his or her own user-item bipartite subgraph and capture the high-order information in a non-existent global graph through collaborative training. However, the existing works on FedGCN are based on a full-precision continuous embedding space, which leads to the high overhead of communication and computation at the clients. In addition, the constructed global graph is usually expensive and suboptimal, and how to construct an accurate and privacy-preserving global graph at a low cost is still a challenge. To address these challenges, we propose a novel privacy-preserving recommendation method, i.e., federated hash graph convolutional network for recommendation (FedHashRec). Specifically, we first design a new hash graph convolutional method (HashRec) to map the full-precision embeddings into the Hamming space and then learn the similarity between users and items. Secondly, we design a privacy-preserving framework with triple protector (TriPro) to construct a fake global graph in the server through efficient client-client and client-server communication. Each item hash code is updated by obtaining the information of the fake user-item bipartite graph to avoid frequent communication. Moreover, we provide a rigorous theoretical proof to show that the information obtained by the server from this fake user-item bipartite graph is lossless. We conduct extensive experiments on four real-world datasets to demonstrate that our method outperforms the existing fully discrete methods in terms of accuracy and efficiency. The datasets, source codes and scripts are available at https://github.com/YannickCodeHome/FedHashRec.
Effective coordination is the cornerstone of success in cooperative Multi-Agent Reinforcement Learning (MARL), where agents work together to achieve a collective goal. However, most existing methods for modeling agent coordination rely primarily on proximity between agents’ features or on incorporating heterogeneity at the behavioral/policy level, while overlooking the valuable diversity present at the observation level. In this paper, we introduce Diametric Coordination Graphs (DiaCoG), a novel framework that dynamically models implicit coordination of agents by integrating both consistency (shared similarities) and discrepancy (unique differences) from agents’ observations, offering a richer understanding of inter-agent relationships. Accordingly, we propose two approaches, DiaCoG-DE and DiaCoG-CE, implementing this framework under two MARL architectures of actor-critic networks, Centralized Training and Decentralized Execution (CTDE) and Centralized Training and Centralized Execution (CTCE), respectively. Through theoretical analyses based on information-theoretic measures, we show that DiaCoG offers superior expressiveness over consistency-based methods for value estimation and action selection. Empirically, we evaluate the proposed methods in the Predator-Prey and Traffic Junction environments, where they outperform baselines in terms of both final average returns/success rates and convergence speed across diverse scenarios. To further showcase the adaptability of DiaCoG across different MARL paradigms, we integrate it with a state-of-the-art value-based approach and compare it against several representative and state-of-the-art methods to demonstrate the enhanced coordination performance by DiaCoG in the StarCraft II Multi-Agent Challenge environment.
