Spatial representation learning is fundamental to GeoAI applications, including urban analytics, as it encodes the shapes, locations, and spatial relationships (topological and distance-based) of geo-entities such as points, polylines, and polygons. Existing methods either target a single geo-entity type or, like Poly2Vec, decompose entities into simpler components to enable Fourier transformation, introducing high computational cost. Moreover, since the transformed space lacks geometric alignment, these methods rely on uniform, non-adaptive sampling, which blurs fine-grained features like edges and boundaries. To address these limitations, we introduce Geo2Vec, a novel method inspired by signed distance fields (SDF) that operates directly in the original space. Geo2Vec adaptively samples points and encodes their signed distances (positive outside, negative inside), capturing geometry without decomposition. A neural network trained to approximate the SDF produces compact, geometry-aware, and unified representations for all geo-entity types. Additionally, we propose a rotation-invariant positional encoding to model high-frequency spatial variations and construct a structured and robust embedding space for downstream GeoAI models. Empirical results show that Geo2Vec consistently outperforms existing methods in representing shape and location, capturing topological and distance relationships, and achieving greater efficiency in real-world GeoAI applications.

Theory of Mind (ToM) refers to the ability to reason about others’ mental states, and higher-order ToM involves considering that others also possess their own ToM. Equipping large language model (LLM)-driven agents with ToM has long been considered to improve their coordination in multiagent collaborative tasks. However, we find that misaligned ToM orders—mismatches in the depth of ToM reasoning between agents—can lead to insufficient or excessive reasoning about others, thereby impairing their coordination. To address this issue, we design an adaptive ToM (A-ToM) agent, which can align in ToM orders with its partner. Based on prior interactions, the agent estimates the partner’s likely ToM order and leverages this estimation to predict the partner’s action, thereby facilitating behavioral coordination. We conduct empirical evaluations on four multi-agent coordination tasks: a repeated matrix game, two grid navigation tasks and an Overcooked task. The results validate our findings on ToM alignment and demonstrate the effectiveness of our AToM agent. Furthermore, we discuss the generalizability of our A-ToM to non-LLM-based agents, as well as what would diminish the importance of ToM alignment.

Stochastic games form the foundational mathematical framework for describing multiagent interactions and underpin the theoretical foundations of multiagent reinforcement learning (MARL) and optimal decision making. However, previous research has typically focused on either two-agent settings or large-scale well-mixed agent populations, where the considered interaction scenarios were far from realistic. In this article, we consider structured populations where agents can interact with immediate neighbors. By using the pair-approximation method, we develop a new dynamical model to describe the $Q$ -learning dynamics in stochastic games on regular graphs. Through comparisons with agent-based simulation results, we validate the accuracy of our dynamical model across various stochastic games, population structures, and algorithm parameters. Our research thus provides both qualitative and quantitative insights into the effects of state transition rules and graph topologies in population dynamics. In particular, we show that, under certain conditions, state transitions can significantly promote the evolution of cooperation in social dilemmas. We also explored the effects of agent degree on cooperation, and unlike previous findings, we show that this can have either positive or negative implications for cooperation depending on the transition rules.

Reciprocity plays a crucial role in maintaining cooperation in human societies and AI systems. In this paper, we focus on reciprocity within multichannel games and examine how cooperation evolves in this context. We propose a unified framework that allows us to evaluate the reputations of interdependent actions across multiple channels while simultaneously exploring both direct and indirect reciprocity mechanisms. We identify partner and semi-partner strategies under both forms of reciprocity, with the former leading to full cooperation and the latter resulting in partial cooperation. Through equilibrium analysis, we characterize the conditions under which full cooperation and partial cooperation emerge. Moreover, we show that when players can link multiple interactions, they learn to coordinate their behavior across different games to maximize overall cooperation. Our findings provide new insights into the maintenance of cooperation across various reciprocity mechanisms and interaction patterns.

Inferring social relationships from mobility data is crucial for many applications because it reflects real-world connections among people. However, large-scale trajectory datasets with ground-truth social ties are exceedingly scarce, making it difficult to train deep models for relationship inference. To address this gap, we propose a transferable social relationship inference framework that can be trained on one high-quality, labeled dataset and then generalized to new datasets, even from different cities. Our framework rests on the key insight that social bonds depend largely on the frequency of individual meetings and the popularity of those meeting locations, both of which can be inferred statistically from raw trajectory data, irrespective of the underlying geographic semantics. It comprises two main modules: 1) Universal Social Relationship Classifier (USRC): A model trained to infer social relationships from trajectory data, and 2) Spatial Embedding Transfer (SET): A location embedding alignment technique that adapts new datasets to the pre-trained USRC model. By aligning location embeddings, SET module enables the pre-trained USRC to interpret previously unseen datasets without extra supervision. Experiments on five public datasets demonstrate that our method achieves state-of-the-art performance in zero-shot social relationship inference, surpassing other unsupervised, and in some cases, even supervised, approaches. Additionally, the SET module significantly improves location embedding alignment, outperforming existing baseline methods. The source code and data are available at https://github.com/chuchen2017/SET.

In this article, we introduce a new theory for payoff control in multichannel learning environments, where agents interact with each other over multiple channels and each channel is a repeated normal form game. We propose two payoff control strategies-partial control and full control-that allow a single agent to set an upper bound to the opponent's expected payoffs summed across all channels, even if the opponent is a reinforcement learning agent. We prove that a partial (or full) control strategy can be obtained by solving a system of inequalities, and characterize the conditions under which such a partial (or full) control strategy exists. We show that by utilizing these control strategies, the agent can influence the opponent's learning evolution and direct it toward a desired viable equilibrium. Our experiments confirm the effectiveness of our theory for payoff control in a wide range of multichannel learning environments.

Understanding and predicting the behavior of large-scale multiagents in games remains a fundamental challenge in multiagent systems. This article examines the role of heterogeneity in equilibrium formation by analyzing how smooth regret matching drives a large number of heterogeneous agents with diverse initial policies toward unified behavior. By modeling the system state as a probability distribution of regrets and analyzing its evolution through the continuity equation, we uncover a key phenomenon in diverse multiagent settings: the variance of the regret distribution diminishes over time, leading to the disappearance of heterogeneity and the emergence of consensus among agents. This universal result enables us to prove convergence to quantal response equilibria in both competitive and cooperative multiagent settings. This work advances the theoretical understanding of multiagent learning and offers a novel perspective on equilibrium selection in diverse game-theoretic scenarios.

Most mobility modeling methods are designed to solve specific tasks, leading to questions regarding their deficiency in generalizability. Inspired by the bloom of foundation models, we proposed a Trajectory Generation framework based on the Diffusion Model (TrajGDM) to capture the universal mobility pattern in a trajectory dataset by learning the trajectory generation process. The process is modeled as a step-by-step uncertainty-reducing process, in which a deep learning network with a novel training method is proposed to learn from the process. We compared the proposed trajectory generation method with six baselines on two public trajectory datasets. The results showed that the similarity between the generated and real trajectory movements measured by the Jensen-Shannon Divergence improved significantly on both datasets. Moreover, we applied zero-shot inferences on two basic trajectory tasks: trajectory prediction and trajectory reconstruction. The accuracy improved by a maximum of 25.6% on two tasks. The universal mobility pattern that is suitable for solving multiple trajectory tasks is verified, inferring the strong generalizability of our model. Finally, the study provides insights into artificial intelligence’s understanding of human mobility by exploring the way the model maps the trajectory in the latent space into reality.

Developing a dynamical model for learning in games has attracted much recent interest. In stochastic games, agents need to make decisions in multiple states, and transitions between states, in turn, influence the dynamics of strategies. While previous works typically focus either on 2-agent stochastic games or on normal form games under an infinite-agent setting, we aim at formally modelling the learning dynamics in stochastic games under the infinite-agent setting. With a novel use of pair-approximation method, we develop a formal model for myopic Q-learning in stochastic games with symmetric state transition. We verify the descriptive power of our model (a partial differential equation) across various games through comparisons with agent-based simulation results. Based on our proposed model, we can gain qualitative and quantitative insights into the influence of transition probabilities on the dynamics of strategies. In particular, we illustrate that a careful design of transition probabilities can help players overcome the social dilemmas and promote cooperation, even if agents are myopic learners.

Capturing the universal movement pattern and simulating human mobility is one of the most important trajectory data-mining tasks. Most of the current mobility modeling methods are specially designed to solve a specific task, which leads to questions regarding generalizability. Aiming to construct a general trajectory foundation model to overcome this weakness, we proposed a generative Trajectory Generation framework based on Diffusion Model (TrajGDM) to capture the universal mobility pattern and simulate human mobility. It is capable of solving multiple trajectory tasks through learning the generation of the trajectory. The generation process of a trajectory is modeled as a step-by-step uncertainty reducing process. A trajectory generator network is proposed to estimate the uncertainty in each step, and a trajectory diffusion and generation process is defined to train the model to simulate the real dataset. Finally, we compared the proposed method with 6 baselines on 2 public trajectory datasets: T-Drive and Geo-life. By comparing 5 different evaluation metrics, the result showed that the similarity between generated and real trajectories' movement character measured by Jensen-Shannon Divergence (JSD) improved by at least 50.3% in both datasets. It also addresses the problem of generating diverse trajectories, which is ignored by most previous models. Moreover, we applied zero-shot inferences on two basic trajectory tasks: trajectory prediction and trajectory reconstruction. The zero-shot prediction accuracy of our model is up to 23.4% higher than the benchmark, and the reconstruction accuracy improves by a maximum of 25.6%.

Modeling the dynamics of multi-agent learning has long been an important research topic. The focus of previous research has been either on 2-agent settings or well-mixed infinitely large agent populations. In this paper, we consider the scenario where n Q-learning agents locate on regular graphs, such that agents can only interact with their neighbors. We examine the local interactions between individuals and their neighbors, and derive a formal model to capture the Q-value dynamics of the entire population. Through comparisons with agent-based simulations on different types of regular graphs, we show that our model describes the agent learning dynamics in an exact manner.

Understanding the learning dynamics in multiagent systems is an important and challenging task. Past research on multi-agent learning mostly focuses on two-agent settings. In this paper, we consider the scenario in which a population of infinitely many agents apply regret minimization in repeated symmetric games. We propose a new formal model based on the master equation approach in statistical physics to describe the evolutionary dynamics in the agent population. Our model takes the form of a partial differential equation, which describes how the probability distribution of regret evolves over time. Through experiments, we show that our theoretical results are consistent with the agent-based simulation results.

Matching buyers with most suitable sellers providing relevant items (e.g., products) is essential for e-commerce platforms to guarantee customer experience. This matching process is usually achieved through modeling inter-group (buyer-seller) proximity by e-commerce ranking systems. However, current ranking systems often match buyers with sellers of various qualities, and the mismatch is detrimental to not only buyers' level of satisfaction but also the platforms' return on investment (ROI). In this paper, we address this problem by incorporating intra-group structural information (e.g., buyer-buyer proximity implied by buyer attributes) into the ranking systems. Specifically, we propose De ep Gr aph E mbe dding (DEGREE), a deep learning based method, to exploit both inter-group and intra-group proximities jointly for structural learning. With a sparse filtering technique, DEGREE can significantly improve the matching performance with computation resources less than that of alternative deep learning based methods. Experimental results demonstrate that DEGREE outperforms state-of-the-art graph embedding methods on real-world e-commence datasets. In particular, our solution boosts the average unit price in purchases during an online A/B test by up to 11.93%, leading to better operational efficiency and shopping experience.

Nowadays, e-commerce has become prevalent world-wide. With the big success of e-commerce, many malicious promotion services also rise: with the goal of increasing sales, malicious merchants attempt to promote their target items by illegally optimizing the search results using fake visits, purchases, etc. In this paper, we study the fraud detection problem on large-scale e-commerce platforms. First, we develop an efficient and scalable AnTi-Fraud system (ATF) to detect e-commerce frauds for large-scale e-commerce platforms, and implement it in parallel on a large-scale computing platform, called Open Data Processing Service (ODPS). Then, we evaluate ATF using two real large-scale e-commerce datasets (with tens of millions users and items). The results demonstrate that both the precision and the recall of ATF can achieve 0.97+, which suggests that ATF is very effective. More importantly, we deploy ATF on the Taobao platform of Alibaba, which is one of the world's largest e-commerce platforms. The evaluation results show that ATF can also achieve an accuracy of 98.16% on Taobao, which again suggests that ATF is very effective and deployable in practice. Our study in this paper is expected to shed light on defending against online frauds for practical e-commerce platforms.

Conducting fraud transactions has become popular among e-commerce sellers to make their products favorable to the platform and buyers, which decreases the utilization efficiency of buyer impressions and jeopardizes the business environment. Fraud detection techniques are necessary but not enough for the platform since it is impossible to recognize all the fraud transactions. In this paper, we focus on improving the platform's impression allocation mechanism to maximize its profit and reduce the sellers' fraudulent behaviors simultaneously. First, we learn a seller behavior model to predict the sellers' fraudulent behaviors from the real-world data provided by one of the largest e-commerce company in the world. Then, we formulate the platform's impression allocation problem as a continuous Markov Decision Process (MDP) with unbounded action space. In order to make the action executable in practice and facilitate learning, we propose a novel deep reinforcement learning algorithm DDPG-ANP that introduces an action norm penalty to the reward function. Experimental results show that our algorithm significantly outperforms existing baselines in terms of scalability and solution quality.