Author name cluster

Jiamou Liu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

42 papers

2 author rows

AAAI Conference 2026 Conference Paper

CLER: Improving Multimodal Financial Reasoning by Cross-MLLM Error Reflection

Shuangyan Deng
Zhongsheng Wang
Rui Mao
Ciprian Doru Giurcăneanu
Jiamou Liu

Recent advances in Multimodal Large Language Models (MLLMs) have enabled joint reasoning over financial textual and visual inputs. However, they still struggle with financial terminology, logical consistency, and numerical computations. Moreover, while commercial large models perform well on reasoning tasks, their high inference costs limit their scalable usage in real world financial applications. We thus propose a cost-effective framework, CLER, that combines contrastive retrieval with step-wise reflection to improve reasoning performance. Also, the reasoning cost is only generated in the test stage when using commercial large models. CLER leverages FinErrorSet, a dataset of 8,000+ mistake correction pairs from diverse open-source MLLMs. A fine grained retriever is trained to identify structurally relevant errors for self-correction through individual reflection. Experiments on three benchmarks show that CLER consistently outperforms other baselines. To our knowledge, CLER is the first framework to use cross-model errors for financial reasoning.

PDF Details DOI

AAMAS Conference 2026 Conference Paper

Equity by Design in Task Allocation: Reverse Auctions with Group and Individual Fairness

Yunxiao Zhao
Mengxiao Zhang
Jiamou Liu
Bakh Khoussainov

Reverse auctions are widely used for budgeted task allocation, but efficiency-only rules can systematically disadvantage higher-cost communities. We ask whether equity can be built into reverseauction task allocation without sacrificing the incentive guarantees. Weintroducetwocomplementaryfairnessnotions: aggregategroup fairness and within-group Lipschitz individual fairness. We propose GIFTA, a two-stage mechanism that learns group-selection probabilities and then allocates within the chosen group using a bid-decreasing randomized rule with truthful payments. We prove that GIFTA is truthful in expectation, individually rational, budgetfeasible, satisfies the Lipschitz guarantee, and achieves asymptotic 𝜖-group fairness; experiments show large disparity reductions with modest social-cost increases.

PDF

AAMAS Conference 2026 Conference Paper

R-Debater: Retrieval-Augmented Debate Generation through Argumentative Memory

Maoyuan Li
Zhongsheng Wang
Haoyuan Li
Jiamou Liu

We present R-Debater, an agentic framework for generating multiturn debates grounded in argumentative memory. Drawing on principles from rhetoric and memory studies, the framework conceptualizes debate as a dynamic process of retrieving and adapting prior arguments to maintain stance consistency, respond to opposing claims, and support assertions with evidence. Specifically, R-Debater integrates a debate knowledge base for retrieving case-like evidence and prior debate moves with a role-based agent that composes coherent utterances across turns. We evaluate R-Debater on two tasks using ORCHID debates: next-utterance generation (assessed by InspireScore) and multi-turn adversarial simulation (evaluated by Debatrix). Our framework outperforms strong LLM baselines in both settings. Human evaluation with 20 experienced debaters further confirms its consistency and evidence use, demonstrating that retrieval grounding combined with structured planning yields more faithful, stance-aligned, and coherent debates. Code and supplementary materials are available at https: //github. com/Maoyuan-li/R-debater.

PDF

AAMAS Conference 2026 Conference Paper

Situational-Constrained Multi-Agent Coordination through Correlated Equilibria

Libo Zhang
Zhirui Zeng
Yang Chen
Jiamou Liu

Correlated Equilibria (CE) provide a robust framework for balancing individual and collective incentives in multi-agent coordination. However, real-world applications often impose situational constraints, i. e. , context-dependent requirements triggered only under specific conditions (e. g. , emergency protocols), which challenge existing methods. To address this, we propose Situational-Constrained Density-Based Correlated Equilibria (SC-DBCE) within a Markov game framework. We develop Situational-Constrained Correlated Policy Iteration (SC-CPI), the first RL algorithm capable of solving these equilibria using a smooth Log-Sum-Exp optimization mechanism. Experiments across multiple scenarios demonstrate that SC-CPI consistently outperforms baselines in both equilibrium quality and constraint adherence.

PDF

AAMAS Conference 2025 Conference Paper

Data Pricing for Graph Neural Networks without Pre-purchased Inspection

Yiping Liu
Mengxiao Zhang
Jiamou Liu
Song Yang

Machine learning (ML) models have become essential tools in various scenarios. Their effectiveness, however, hinges on a substantial volume of data for satisfactory performance. Model marketplaces have thus emerged as crucial platforms bridging model consumers seeking ML solutions and data owners possessing valuable data. These marketplaces leverage model trading mechanisms to properly incentive data owners to contribute their data, and return a well performing ML model to the model consumers. However, existing model trading mechanisms often assume the data owners are willing to share their data before being paid, which is not reasonable in real world. Given that, we propose a novel mechanism, named Structural Importance based Model Trading (SIMT) mechanism, that assesses the data importance and compensates data owners accordingly without disclosing the data. Specifically, SIMT procures feature and label data from data owners according to their structural importance, and then trains a graph neural network for model consumers. Theoretically, SIMT ensures incentive compatible, individual rational and budget feasible. The experiments on five popular datasets validate that SIMT consistently outperforms vanilla baselines by up to 40% in both MacroF1 and MicroF1.

PDF

AAAI Conference 2025 Conference Paper

Learning Verified Safe Neural Network Controllers for Multi-Agent Path Finding

Mingyue Zhang
Nianyu Li
Yi Chen
Jialong Li
Xiao-Yi Zhang
Hengjun Zhao
Jiamou Liu
Wu Chen

Multi-agent path finding (MAPF) is a safety-critical scenario where the goal is to secure collision-free trajectories from initial to desired locations. However, due to system complexity and uncertainty, integrating learning-based controllers with MAPF is challenging and cannot theoretically guarantee the safety of the learned controllers. In response, our study proposes a verified safe multi-agent neural control (VSMANC) approach for MAPF, focusing on the unified training of Decentralized Control Barrier Functions (DCBF) and controllers to enhence safety. VSMANC enables all agents to concurrently learn controllers and DCBFs using a unified loss function designed to maximize safety, adhere to standard control policies, and incorporate path-finding-related heuristics. We also propose a formal verification-guided retraining process to both verify the properties of the learned DCBFs and generate counterexamples for retraining, thereby providing a verified safety guarantee. We validate our approach through shape formation experiments and UAV simulations, demonstrating significant improvements in safety and effectiveness in complex multi-agent environments.

PDF Details DOI

AAMAS Conference 2025 Conference Paper

SFedRec: A Federated Learning Framework for Dynamic Session-based Recommendation

Hexiao Zhang
Yanni Tang
Jiamou Liu
Wu Chen

Session-based recommendation systems are critical for capturing users’ evolving interests in real-time interactions. However, applying such systems in a federated learning (FL) setting presents challenges related to decentralized data and privacy preservation. To address this, we propose SFedRec, a session-based federated recommendation framework that integrates long-term user preferences with dynamicd session-based behaviors. SFedRec builds decentralized heterogeneous knowledge graphs to model user-item interactions and social connections, utilizing a graph neural network to learn user representations while ensuring privacy through Local Differential Privacy (LDP). Extensive experiments on three real-world datasets demonstrate that SFedRec outperforms stateof-the-art federated recommendation models, showing significant improvements in both general and cold-start scenarios.

PDF

IJCAI Conference 2025 Conference Paper

Situational-Constrained Sequential Resources Allocation via Reinforcement Learning

Libo Zhang
Yang Chen
Toru Takisaka
Kaiqi Zhao
Weidong Li
Jiamou Liu

Sequential Resource Allocation with situational constraints presents a significant challenge in real-world applications, where resource demands and priorities are context-dependent. This paper introduces a novel framework, SCRL, to address this problem. We formalize situational constraints as logic implications and develop a new algorithm that dynamically penalizes constraint violations. To handle situational constraints effectively, we propose a probabilistic selection mechanism to overcome limitations of traditional constraint reinforcement learning (CRL) approaches. We evaluate SCRL across two scenarios: medical resource allocation during a pandemic and pesticide distribution in agriculture. Experiments demonstrate that SCRL outperforms existing baselines in satisfying constraints while maintaining high resource efficiency, showcasing its potential for real-world, context-sensitive decision-making tasks.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Trust Region Reward Optimization and Proximal Inverse Reward Optimization Algorithm

Yang Chen
Menglin Zou
Jiaqi Zhang
Yitan Zhang
Junyi Yang
Gaël Gendron
Libo Zhang
Jiamou Liu

Inverse Reinforcement Learning (IRL) learns a reward function to explain expert demonstrations. Modern IRL methods often use the adversarial (minimax) formulation that alternates between reward and policy optimization, which often lead to {\em unstable} training. Recent non-adversarial IRL approaches improve stability by jointly learning reward and policy via energy-based formulations but lack formal guarantees. This work bridges this gap. We first present a unified view showing canonical non-adversarial methods explicitly or implicitly maximize the likelihood of expert behavior, which is equivalent to minimizing the expected return gap. This insight leads to our main contribution: Trust Region Reward Optimization (TRRO), a framework that guarantees monotonic improvement in this likelihood via a Minorization-Maximization process. We instantiate TRRO into Proximal Inverse Reward Optimization (PIRO), a practical and stable IRL algorithm. Theoretically, TRRO provides the IRL counterpart to the stability guarantees of Trust Region Policy Optimization (TRPO) in forward RL. Empirically, PIRO matches or surpasses state-of-the-art baselines in reward recovery, policy imitation with high sample efficiency on MuJoCo and Gym-Robotics benchmarks and a real-world animal behavior modeling task.

PDF Details

ECAI Conference 2024 Conference Paper

Balancing Efficiency with Equality: Auction Design with Group Fairness Concerns

Fengjuan Jia
Mengxiao Zhang
Jiamou Liu
Bakhadyr Khoussainov

The issue of fairness in AI arises from discriminatory practices in applications like job recommendations and risk assessments, emphasising the need for algorithms that do not discriminate based on group characteristics. This concern is also pertinent to auctions, commonly used for resource allocation, which necessitate fairness considerations. Our study examines auctions with groups distinguished by specific attributes, seeking to (1) define a fairness notion that ensures equitable treatment for all, (2) identify mechanisms that adhere to this fairness while preserving incentive compatibility, and (3) explore the balance between fairness and seller’s revenue. We introduce two fairness notions—group fairness and individual fairness—and propose two corresponding auction mechanisms: the Group Probability Mechanism, which meets group fairness and incentive criteria, and the Group Score Mechanism, which also encompasses individual fairness. Through experiments, we validate these mechanisms’ effectiveness in promoting fairness and examine their implications for seller revenue.

Details

AAMAS Conference 2024 Conference Paper

Behaviour Modelling of Social Animals via Causal Structure Discovery and Graph Neural Networks

Gaël Gendron
Yang Chen
Mitchell Rogers
Yiping Liu
Mihailo Azhar
Shahrokh Heidari
David Arturo Soriano Valdez
Kobe Knowles

Better understanding the natural world is a crucial task with a wide range of applications. In environments with close proximity between humans and animals, such as zoos, it is essential to better understand the causes behind animal behaviour to predict unusual changes, mitigate their detrimental effects and increase the well-being of animals. However, the complex social behaviours of mammalian groups remain largely unexplored. In this work, we propose a method to build behavioural models using causal structure discovery and graph neural networks for time series. We apply this method to a mob of meerkats in a zoo environment and study its ability to predict future actions and model the behaviour distribution at an individual-level and at a group level. We show that our method can match and outperform standard deep learning architectures and generate more realistic data, while using fewer parameters and providing increased interpretability.

PDF

AAMAS Conference 2024 Conference Paper

Memory-Based Resilient Control Against Non-cooperation in Multi-agent Flocking

Mingyue Zhang
Nianyu Li
Jialong Li
Jiachun Liao
Jiamou Liu

Inspired by natural flocking behaviors, researchers aim to develop a distributed control approach for artificial agents to mimic these behaviors. The main challenge lies in maintaining the resilience of the artificial flock, as some agents inevitably display non-cooperative behavior, thereby deviating from the flocking objective. Existing control approaches, especially those based on learning algorithm, are susceptible to forgetting issues that non-cooperative agents can exploit to disrupt the flock formation. To address this problem, this study introduces a memory-based resilient control approach that strategically analyzes historical data across three distinct time scales (long, short, and periodic). The implementation of a long short periodic-term memory (LSP) algorithm employs accumulative discounted credibility evaluated by Q-learning to recognize long-term non-cooperation, utilizes a filtering rule to establish a trusted set excluding short-term non-cooperation, and integrates fast Fourier transform to refine the trusted set against periodic inconsistency. We assess the effectiveness of this approach through extensive experiments. The results highlight the potential and advantages of using LSP in flocking, enhancing the resilience of multiagent flocking against complex non-cooperative threats.

PDF

AAAI Conference 2024 Conference Paper

Meta-Inverse Reinforcement Learning for Mean Field Games via Probabilistic Context Variables

Yang Chen
Xiao Lin
Bo Yan
Libo Zhang
Jiamou Liu
Neset Özkan Tan
Michael Witbrock

Designing suitable reward functions for numerous interacting intelligent agents is challenging in real-world applications. Inverse reinforcement learning (IRL) in mean field games (MFGs) offers a practical framework to infer reward functions from expert demonstrations. While promising, the assumption of agent homogeneity limits the capability of existing methods to handle demonstrations with heterogeneous and unknown objectives, which are common in practice. To this end, we propose a deep latent variable MFG model and an associated IRL method. Critically, our method can infer rewards from different yet structurally similar tasks without prior knowledge about underlying contexts or modifying the MFG model itself. Our experiments, conducted on simulated scenarios and a real-world spatial taxi-ride pricing problem, demonstrate the superiority of our approach over state-of-the-art IRL methods in MFGs.

PDF Details DOI

ECAI Conference 2024 Conference Paper

Meta-Mechanisms for Combinatorial Auctions over Social Networks

Yuan Fang
Mengxiao Zhang
Jiamou Liu
Bakhadyr Khoussainov

Recently there has been a large amount of research designing mechanisms for auction scenarios where the bidders are connected in a social network. Different from the existing studies in this field that focus on specific auction scenarios e. g. single-unit auction and multi-unit auction, this paper considers the following question: is it possible to design a scheme that, given a classical auction scenario and a mechanism M˜ suited for it, produces a mechanism in the network setting that preserves the key properties of M˜? To answer this question, we design meta-mechanisms that provide a uniform way of transforming mechanisms from classical models to mechanisms over networks and prove that the desirable properties are preserved by our meta-mechanisms. Our meta-mechanisms provide solutions to combinatorial auction scenarios in the network setting: (1) combinatorial auction with single-minded buyers and (2) combinatorial auction with general monotone valuation. To the best of our knowledge, this is the first work that designs combinatorial auctions over a social network.

Details

AAMAS Conference 2023 Conference Paper

Adversarial Inverse Reinforcement Learning for Mean Field Games

Yang Chen
Libo Zhang
Jiamou Liu
Michael Witbrock

Goal-based agents respond to environments and adjust behaviour accordingly to reach objectives. Understanding incentives of interacting agents from observed behaviour is a core problem in multi-agent systems. Inverse reinforcement learning (IRL) solves this problem, which infers underlying reward functions by observing the behaviour of rational agents. Despite IRL being principled, it becomes intractable when the number of agents grows because of the curse of dimensionality and the explosion of agent interactions. The formalism of Mean field games (MFGs) has gained momentum as a mathematically tractable paradigm for studying large-scale multi-agent systems. By grounding IRL in MFGs, recent research attempts to push the limits of the agent number in IRL. However, the study of IRL for MFGs is far from being mature as existing methods assume strong rationality, while real-world agents often exhibit bounded rationality due to the limited cognitive or computational capacity. Towards a more general and practical IRL framework for MFGs, this paper proposes Mean-Field Adversarial IRL, a novel framework capable of tolerating bounded rationality. We build it upon the maximum entropy principle, adversarial learning, and a new equilibrium concept for MFGs. We evaluate our machinery on simulated tasks with imperfect demonstrations resulting from bounded rationality. Experimental results demonstrate the superiority of MF-AIRL over existing methods in reward recovery.

PDF

AAMAS Conference 2023 Conference Paper

Differentially Private Diffusion Auction: The Single-unit Case

Fengjuan Jia
Mengxiao Zhang
Jiamou Liu
Bakh Khoussainov

Diffusion auction refers to an emerging paradigm where an auctioneer utilises a social network to attract potential buyers. We consider the risks of disclosing sensitive preferences of buyers from the published auction outcome and initiate the study of differential privacy in diffusion auction. We study the single-unit case and design two differentially private diffusion mechanisms (DPDMs): recursive DPDM and layered DPDM. We prove their incentive and privacy properties, and then empirically compare their performance on real and synthetic datasets.

PDF

UAI Conference 2023 Conference Paper

Incentivising Diffusion while Preserving Differential Privacy

Fengjuan Jia
Mengxiao Zhang
Jiamou Liu
Bakhadyr Khoussainov

Diffusion auction refers to an emerging paradigm of online marketplace where an auctioneer utilises a social network to attract potential buyers. Diffusion auction poses significant privacy risks. From the auction outcome, it is possible to infer hidden, and potentially sensitive, preferences of buyers. To mitigate such risks, we initiate the study of differential privacy (DP) in diffusion auction mechanisms. DP is a well-established notion of privacy that protects a system against inference attacks. Achieving DP in diffusion auctions is non-trivial as the well-designed auction rules are required to incentivise the buyers to truthfully report their neighbourhood. We study the single-unit case and design two differentially private diffusion mechanisms (DPDMs): recursive DPDM and layered DPDM. We prove that these mechanisms guarantee differential privacy, incentive compatibility and individual rationality for both valuations and neighbourhood. We then empirically compare their performance on real and synthetic datasets.

Details

ECAI Conference 2023 Conference Paper

Integrated Private Data Trading Systems for Data Marketplaces

Weidong Li
Mengxiao Zhang
Libo Zhang 0006
Jiamou Liu

In the digital age, data is a valuable commodity, and data marketplaces offer lucrative opportunities for data owners to monetize their private data. However, data privacy is a significant concern, and differential privacy has become a popular solution to address this issue. Private data trading systems (PDQS) facilitate the trade of private data by determining which data owners to purchase data from, the amount of privacy purchased, and providing specific aggregation statistics while protecting the privacy of data owners. However, existing PDQS with separated procurement and query processes are prone to over-perturbation of private data and lack trustworthiness. To address this issue, this paper proposes a framework for PDQS with an integrated procurement and query process to avoid excessive perturbation of private data. We also present two instances of this framework, one based on a greedy approach and another based on a neural network. Our experimental results show that both of our mechanisms outperformed the separately conducted procurement and query mechanism under the same budget regarding accuracy.

Details

AAMAS Conference 2023 Conference Paper

Learning Density-Based Correlated Equilibria for Markov Games

Libo Zhang
Yang Chen
Toru Takisaka
Bakh Khoussainov
Michael Witbrock
Jiamou Liu

Correlated Equilibrium (CE) is a well-established solution concept that captures coordination among agents and enjoys good algorithmic properties. In real-world multi-agent systems, in addition to being in equilibrium, agents’ policies are often expected to meet requirements with respect to safety, and fairness. Such additional requirements can often be expressed in terms of the state density which measures the state-visitation frequencies during the course of a game. However, existing CE notions or CE-finding approaches cannot explicitly specify a CE with particular properties concerning state density; they do so implicitly by either modifying reward functions or using value functions as the selection criteria. The resulting CE may thus not fully fulfil the state-density requirements. In this paper, we propose Density-Based Correlated Equilibria (DBCE), a new notion of CE that explicitly takes state density as a selection criterion. Concretely, we instantiate DBCE by specifying different state-density requirements motivated by real-world applications. To compute DBCE, we put forward the Density Based Correlated Policy Iteration algorithm for the underlying control problem. We perform experiments on various games where results demonstrate the advantage of our CE-finding approach over existing methods in scenarios with state-density concerns.

PDF

AAAI Conference 2023 Conference Paper

MSDC: Exploiting Multi-State Power Consumption in Non-intrusive Load Monitoring Based on a Dual-CNN Model

Jialing He
Jiamou Liu
Zijian Zhang
Yang Chen
Yiwei Liu
Bakh Khoussainov
Liehuang Zhu

Non-intrusive load monitoring (NILM) aims to decompose aggregated electrical usage signal into appliance-specific power consumption and it amounts to a classical example of blind source separation tasks. Leveraging recent progress on deep learning techniques, we design a new neural NILM model {\em Multi-State Dual CNN} (MSDC). Different from previous models, MSDC explicitly extracts information about the appliance's multiple states and state transitions, which in turn regulates the prediction of signals for appliances. More specifically, we employ a dual-CNN architecture: one CNN for outputting state distributions and the other for predicting the power of each state. A new technique is invented that utilizes conditional random fields (CRF) to capture state transitions. Experiments on two real-world datasets REDD and UK-DALE demonstrate that our model significantly outperform state-of-the-art models while having good generalization capacity, achieving 6%-10% MAE gain and 33%-51% SAE gain to unseen appliances.

PDF Details DOI

ECAI Conference 2023 Conference Paper

Multi-Unit Auction over a Social Network

Yuan Fang
Mengxiao Zhang
Jiamou Liu
Bakhadyr Khoussainov
Mingyu Xiao 0001

Diffusion auction is an emerging business model where a seller aims to incentivise buyers in a social network to diffuse the auction information thereby attracting potential buyers. We focus on designing mechanisms for multi-unit diffusion auctions. Despite numerous attempts at this problem, existing mechanisms either fail to be incentive compatible (IC) or achieve only an unsatisfactory level of social welfare (SW). Here, we propose a novel graph exploration technique to realise multi-item diffusion auction. This technique ensures that potential competition among buyers stay “localised” so as to facilitate truthful bidding. Using this technique, we design multi-unit diffusion auction mechanisms MUDAN and MUDAN-m. Both mechanisms satisfy, among other properties, IC and 1/m-weak efficiency. We also show that they achieve optimal social welfare for the class of rewardless diffusion auctions. While MUDAN addresses the bottleneck case when each buyer demands only a single item, MUDAN-m handles the more general, multi-demand setting. We further demonstrate that these mechanisms achieve near-optimal social welfare through experiments.

Details

AAAI Conference 2023 Conference Paper

USER: Unsupervised Structural Entropy-Based Robust Graph Neural Network

Yifei Wang
Yupan Wang
Zeyu Zhang
Song Yang
Kaiqi Zhao
Jiamou Liu

Unsupervised/self-supervised graph neural networks (GNN) are susceptible to the inherent randomness in the input graph data, which adversely affects the model's performance in downstream tasks. In this paper, we propose USER, an unsupervised and robust version of GNN based on structural entropy, to alleviate the interference of graph perturbations and learn appropriate representations of nodes without label information. To mitigate the effects of undesirable perturbations, we analyze the property of intrinsic connectivity and define the intrinsic connectivity graph. We also identify the rank of the adjacency matrix as a crucial factor in revealing a graph that provides the same embeddings as the intrinsic connectivity graph. To capture such a graph, we introduce structural entropy in the objective function. Extensive experiments conducted on clustering and link prediction tasks under random-perturbation and meta-attack over three datasets show that USER outperforms benchmarks and is robust to heavier perturbations.

PDF Details DOI

AAMAS Conference 2022 Conference Paper

Individual-Level Inverse Reinforcement Learning for Mean Field Games

Yang Chen
Libo Zhang
Jiamou Liu
Shuyue Hu

The recent mean field game (MFG) formalism has enabled the application of inverse reinforcement learning (IRL) methods in largescale multi-agent systems, with the goal of inferring reward signals that can explain demonstrated behaviours of large populations. The existing IRL methods for MFGs are built upon reducing an MFG to a Markov decision process (MDP) defined on the collective behaviours and average rewards of the population. However, this paper reveals that the reduction from MFG to MDP holds only for the fully cooperative setting. This limitation invalidates existing IRL methods on MFGs with non-cooperative environments. To measure more general behaviours in large populations, we study the use of individual behaviours to infer ground-truth reward functions for MFGs. We propose Mean Field IRL (MFIRL), the first dedicated IRL framework for MFGs that can handle both cooperative and non-cooperative environments. Based on this theoretically justified framework, we develop a practical algorithm effective for MFGs with unknown dynamics. We evaluate MFIRL on both cooperative and mixed cooperative-competitive scenarios with many agents. Results demonstrate that MFIRL excels in reward recovery, sample efficiency and robustness in the face of changing dynamics.

PDF

NeSy Conference 2022 Conference Paper

Multi-Step Deductive Reasoning Over Natural Language: An Empirical Study on Out-of-Distribution Generalisation

Qiming Bao 0001
Alex Yuxuan Peng
Tim Hartill
Neset Tan
Zhenyun Deng
Michael Witbrock
Jiamou Liu

Combining deep learning with symbolic logic reasoning aims to capitalize on the success of both fields and is drawing increasing attention. Inspired by DeepLogic, an end-to-end model trained to perform inference on logic programs, we introduce IMA-GloVe-GA, an iterative neural inference network for multi-step reasoning expressed in natural language. In our model, reasoning is performed using an iterative memory neural network based on RNN with a gate attention mechanism. We evaluate IMA-GloVe-GA on three datasets: PARARULES, CONCEPTRULES V1 and CONCEPTRULES V2. Experimental results show DeepLogic with gate attention can achieve higher test accuracy than DeepLogic and other RNN baseline models. Our model achieves better out-of-distribution generalisation than RoBERTa-Large when the rules have been shuffled. Furthermore, to address the issue of unbalanced distribution of reasoning depths in the current multi-step reasoning datasets, we develop PARARULE-Plus, a large dataset with more examples that require deeper reasoning steps. Experimental results show that the addition of PARARULEPlus can increase the model’s performance on examples requiring deeper reasoning depths. The source code and data are available at https: //github. com/Strong-AI-Lab/Multi-Step-Deductive-Reasoning-OverNatural-Language.

Details

AAMAS Conference 2022 Conference Paper

Residual Entropy-based Graph Generative Algorithms

Wencong Liu
Jiamou Liu
Zijian Zhang
Yiwei Liu
Liehuang Zhu

Classification and clustering are crucial tasks that recognize the identities and the communities of nodes in a graph. Several methods have been proposed to reduce the accuracy of node classification and clustering through graph neural networks (GNN). Existing defense methods usually modify the model architecture and adopt countermeasure training to enhance the robustness of the node classification and clustering. However, these defense methods are model-oriented and not robust. To alleviate the problem, this paper first proposes a robust node classification metric based on residual entropy. More concretely, we prove that maximizing the residual entropy helps to improve the robustness of the classification accuracy. We them propose two graph generative algorithms to resist against two kinds of GNN-based attacks, the untargeted and the targeted attacks. Finally, experimental analysis show that the proposed algorithms outperform the existing defense works under five classic datasets. 1

PDF

ICML Conference 2021 Conference Paper

From Local to Global Norm Emergence: Dissolving Self-reinforcing Substructures with Incremental Social Instruments

Yiwei Liu
Jiamou Liu
Kaibin Wan
Zhan Qin
Zijian Zhang 0001
Bakhadyr Khoussainov
Liehuang Zhu

Norm emergence is a process where agents in a multi-agent system establish self-enforcing conformity through repeated interactions. When such interactions are confined to a social topology, several self-reinforcing substructures (SRS) may emerge within the population. This prevents a formation of a global norm. We propose incremental social instruments (ISI) to dissolve these SRSs by creating ties between agents. Establishing ties requires some effort and cost. Hence, it is worth to design methods that build a small number of ties yet dissolve the SRSs. By using the notion of information entropy, we propose an indicator called the BA-ratio that measures the current SRSs. We find that by building ties with minimal BA-ratio, our ISI is effective in facilitating the global norm emergence. We explain this through our experiments and theoretical results. Furthermore, we propose the small-degree principle in minimising the BA-ratio that helps us to design efficient ISI algorithms for finding the optimal ties. Experiments on both synthetic and real-world network topologies demonstrate that our adaptive ISI is efficient at dissolving SRS.

Details

TAAS Journal 2021 Journal Article

Gist Trace-based Learning: Efficient Convention Emergence from Multilateral Interactions

Shuyue Hu
Chin-wing Leung
Ho-fung Leung
Jiamou Liu

The concept of conventions has attracted much attention in the multi-agent system research. In this article, we study the emergence of conventions from repeated n -player coordination games. Distributed agents learn their policies independently and are capable of observing their neighbours in a network topology. We distinguish two types of information representation about the observations: gist trace and verbatim trace. We conjecture that learning based on the gist trace, which overlooks the details and focuses only on the general choice of action of a neighbourhood, should achieve efficient convention emergence. To this end, a novel learning method that makes use of the gist trace is proposed. The experimental results confirm that the proposed method establishes conventions much faster than the state-of-the-art learning methods across diverse settings of multi-agent systems. In particular, the use of gist trace derived at a low level of abstraction further improves the efficiency of convention emergence.

Details DOI

UAI Conference 2020 Conference Paper

Selling Data at an Auction under Privacy Constraints

Mengxiao Zhang
Fernando Beltrán 0001
Jiamou Liu

Private data query combines mechanism design with privacy protection to produce aggregated statistics from privately-owned data records. The problem arises in a data marketplace where data owners have personalised privacy requirements and private data valuations. We focus on the case when the data owners are single-minded, i. e. , they are willing to release their data only if the data broker guarantees to meet their announced privacy requirements. For a data broker who wants to purchase data from such data owners, we propose the SingleMindedQuery (SMQ) mechanism, which uses a reverse auction to select data owners and determine compensations. SMQ satisfies interim incentive compatibility, individual rationality, and budget feasibility. Moreover, it uses purchased privacy expectation maximisation as a principle to produce accurate outputs for commonly-used queries such as counting, median and linear predictor. The effectiveness of our method is empirically validated by a series of experiments.

Details

AAMAS Conference 2019 Conference Paper

On the Maximization of Influence Over an Unknown Social Network

Bo Yan
Kexiu Song
Jiamou Liu
Fanku Meng
Yiping Liu
Hongyi Su

Influence maximization is a well-investigated problem which asks for key individuals who have significant influence in a given social network. This paper addresses this problem when the social network structure is hidden. We adopt the framework of influence learning from samples and build a neural network model to represent the information diffusion process. Based on the model, we propose two new algorithms NeuGreedy and NeuMax. NeuGreedy simulates the traditional greedy algorithm whilst NeuMax utilizes the weights of connections between neurons. We test the algorithms on both synthetic and real-world datasets. The results verify the effectiveness of the proposed methods as compared to existing algorithms with or without the network structure.

PDF

NeurIPS Conference 2019 Conference Paper

REM: From Structural Entropy to Community Structure Deception

Yiwei Liu
Jiamou Liu
Zijian Zhang
Liehuang Zhu
Angsheng Li

This paper focuses on the privacy risks of disclosing the community structure in an online social network. By exploiting the community affiliations of user accounts, an attacker may infer sensitive user attributes. This raises the problem of community structure deception (CSD), which asks for ways to minimally modify the network so that a given community structure maximally hides itself from community detection algorithms. We investigate CSD through an information-theoretic lens. To this end, we propose a community-based structural entropy to express the amount of information revealed by a community structure. This notion allows us to devise residual entropy minimization (REM) as an efficient procedure to solve CSD. Experimental results over 9 real-world networks and 6 community detection algorithms show that REM is very effective in obfuscating the community structure as compared to other benchmark methods.

PDF Details

AAMAS Conference 2019 Conference Paper

To be Big Picture Thinker or Detail-Oriented? Utilizing Perceived Gist Information to Achieve Efficient Convention Emergence with Bilateralism and Multilateralism

Shuyue Hu
Chin-wing Leung
Ho-fung Leung
Jiamou Liu

Recently, the study of social conventions (or norms) has attracted much attention. In this paper, we study the emergence of conventions from agents’ repeated coordination games via bilateralism and multilateralism. We assume that agents can perceive the gist information, i. e. , a big picture of how popular each action is in their neighbourhood. A novel reinforcement learning approach which utilizes the gist information is proposed. Experiment verifies that the proposed approach significantly outperforms the baseline and the state-of-the-art approaches, in terms of the speed of convention emergence.

PDF

AAMAS Conference 2018 Conference Paper

Balancing the Pain and Gain of Hobnobbing: Utility-Based Network Building over Atributed Social Networks

Yijin Cai
Hong Zheng
Jiamou Liu
Bo Yan
Hongyi Su
Yiping Liu

The establishment of interpersonal ties is a pivotal problem in the structural analysis of social networks. In particular, link recommendation problem asks for valuable future links to establish by an individual. Existing methods for this problem rely on link prediction that evaluates the likelihood of successful tie creation between two individuals. Such methods do not consider the social capital gained by agents, nor do they concern with the required cost of this process. In light of this limitation, we propose a utility-based network building problem, with an aim to strike a balance between the gained social capital – in the form of closeness centrality – and the cost of establishing ties. We propose algorithms to solve this problem over networks whose nodes may or may not be labelled with attributes, and test their performance on a range of synthesized and real-world social networks. By having multiple agents adopting utility-based network building strategies, we propose a suite of models of network formation and demonstrate empirically that the they capture important structural properties. In particular, we investigate the emergence of a core/periphery structure as a joint result of preferential attachment and network building strategies.

PDF

IJCAI Conference 2018 Conference Paper

From the Periphery to the Core: Information Brokerage in an Evolving Network

Bo Yan
Yiping Liu
Jiamou Liu
Yijin Cai
Hongyi Su
Hong Zheng

Interpersonal ties are pivotal to individual efficacy, status and performance in an agent society. This paper explores three important and interrelated themes in social network theory: the center/periphery partition of the network; network dynamics; and social integration of newcomers. We tackle the question: How would a newcomer harness information brokerage to integrate into a dynamic network going from periphery to center? We model integration as the interplay between the newcomer and the dynamics network and capture information brokerage using a process of relationship building. We analyze theoretical guarantees for the newcomer to reach the center through tactics; proving that a winning tactic always exists for certain types of network dynamics. We then propose three tactics and show their superior performance over alternative methods on four real-world datasets and four network models. In general, our tactics place the newcomer to the center by adding very few new edges on dynamic networks with ~14000 nodes.

PDF Details

AAAI Conference 2017 Conference Paper

Network, Popularity and Social Cohesion: A Game-Theoretic Approach

Jiamou Liu
Ziheng Wei

In studies of social dynamics, cohesion refers to a group’s tendency to stay in unity, which – as argued in sociometry – arises from the network topology of interpersonal ties. We follow this idea and propose a game-based model of cohesion that not only relies on the social network, but also re- ﬂects individuals’ social needs. In particular, our model is a type of cooperative games where players may gain popularity by strategically forming groups. A group is socially cohesive if the grand coalition is core stable. We study social cohesion in some special types of graphs and draw a link between social cohesion and the classical notion of structural cohesion (White and Harary 2001). We then focus on the problem of deciding whether a given social network is socially cohesive and show that this problem is CoNP-complete. Nevertheless, we give two efﬁcient heuristics for coalition structures where players enjoy high popularity and experimentally evaluate their performances.

PDF Details

AAMAS Conference 2017 Conference Paper

What Becomes of the Broken Hearted? : An Agent-Based Approach to Self-Evaluation, Interpersonal Loss, and Suicide Ideation

Jiamou Liu
Luqi Li
Kyle Russell

Social surroundings greatly affect how we perceive ourselves. Examining the network of social relations may help us reveal important information about our psyche. We build on this idea and propose a model that is consistent with social psychological theories and connects individuals’ emotional states with their social relations. The model consists of a social network of agents. Using a centrality-based measure of self-evaluation, the model quantifies the amount of interpersonal loss experienced by agents as their social relations change. Applying this model, we analyze interpersonal loss of agents in standard network structures under different conditions. We then draw a link to suicide and discuss two real-world suicide incidents. Finally, we simulate dynamics of large random networks and investigate how network structures affect suicide ideation and its possible cascades.

PDF

IJCAI Conference 2016 Conference Paper

How to Build Your Network? A Structural Analysis

Anastasia Moskvina
Jiamou Liu

Creating new ties in a social network facilitates knowledge exchange and affects positional advantage. In this paper, we study the process, which we call network building, of establishing ties between two existing social networks in order to reach certain structural goals. We focus on the case when one of the two networks consists only of a single member and motivate this case from two perspectives. The first perspective is socialization: we ask how a newcomer can forge relationships with an existing network to place herself at the center. We prove that obtaining optimal solutions to this problem is NP-complete, and present several efficient algorithms to solve this problem and compare them with each other. The second perspective is network expansion: we investigate how a network may preserve or reduce its diameter through linking with a new node, hence ensuring small distance between its members. For both perspectives the experiment demonstrates that a small number of new links is usually sufficient to reach the respective goal.

PDF Details

TCS Journal 2015 Journal Article

Dynamic algorithms for monotonic interval scheduling problem

Alexander Gavruskin
Bakhadyr Khoussainov
Mikhail Kokho
Jiamou Liu

We investigate dynamic algorithms for the interval scheduling problem. We focus on the case when the set of intervals is monotonic. This is when no interval properly contains another interval. We provide two data structures for representing the intervals that allow efficient insertion, removal and various query operations. The first dynamic algorithm, based on the data structure called compatibility forest, runs in amortised time O ( log 2 ⁡ n ) for insertion and removal and O ( log ⁡ n ) for query. The second dynamic algorithm, based on the data structure called linearised tree, runs in time O ( log ⁡ n ) for insertion, removal and query. We discuss differences and similarities of these two data structures through theoretical and experimental results.

Details DOI

CSL Conference 2015 Conference Paper

Infinite and Bi-infinite Words with Decidable Monadic Theories

Dietrich Kuske
Jiamou Liu
Anastasia Moskvina

We study word structures of the form (D, <=, P) where D is either N or Z, <= is a linear ordering on D and P in D is a predicate on D. In particular we show: (a) The set of recursive omega-words with decidable monadic second order theories is Sigma_3-complete. (b) We characterise those sets P subset of Z that yield bi-infinite words (Z, <=, P) with decidable monadic second order theories. (c) We show that such "tame" predicates P exist in every Turing degree. (d) We determine, for P subset of Z, the number of predicates Q subset of Z such that (Z, <=, P) and (Z, <=, Q) are indistinguishable. Through these results we demonstrate similarities and differences between logical properties of infinite and bi-infinite words.

Details

MFCS Conference 2013 Conference Paper

A Polychromatic Ramsey Theory for Ordinals

Martin Huschenbett
Jiamou Liu

Abstract The Ramsey degree of an ordinal α is the least number n such that any colouring of the edges of the complete graph on α using finitely many colours contains an n -chromatic clique of order type α. The Ramsey degree exists for any ordinal α < ω ω. We provide an explicit expression for computing the Ramsey degree given α. We further establish a version of this result for automatic structures. In this version the ordinal and the colouring are presentable by finite automata and the clique is additionally required to be regular. The corresponding automatic Ramsey degree turns out to be greater than the set theoretic Ramsey degree. Finally, we demonstrate that a version for computable structures fails.

Details

TCS Journal 2011 Journal Article

Deciding the isomorphism problem in classes of unary automatic structures

Jiamou Liu
Mia Minnes

We solve the isomorphism problem for certain classes of unary automatic structures: unary automatic equivalence relations, unary automatic linear orders, and unary automatic trees. That is, we provide algorithms which decide whether two given elements of these classes are isomorphic. In doing so, we define new finite representations for these structures which give normal forms. 1 1 A related paper which focuses on time and space complexity for these unary automatic structures has been submitted for publication by the authors elsewhere.

Details DOI

CSL Conference 2010 Conference Paper

The Isomorphism Problem for omega -Automatic Trees

Dietrich Kuske
Jiamou Liu
Markus Lohrey

Abstract The main result of this paper is that the isomorphism problem for ω -automatic trees of finite height is at least as hard as second-order arithmetic and therefore not analytical. This strengthens a recent result by Hjorth, Khoussainov, Montalbán, and Nies [9] showing that the isomorphism problem for ω -automatic structures is not \(\Sigma^1_2\). Moreover, assuming the continuum hypothesis CH, we can show that the isomorphism problem for ω -automatic trees of finite height is recursively equivalent with second-order arithmetic. On the way to our main results, we show lower and upper bounds for the isomorphism problem for ω -automatic trees of every finite height: (i) It is decidable ( \(\Pi^0_1\) -complete, resp.) for height 1 (2, resp.), (ii) \(\Pi^1_1\) -hard and in \(\Pi^1_2\) for height 3, and (iii) \(\Pi^1_{n-3}\) - and \(\Sigma^1_{n-3}\) -hard and in \(\Pi^1_{2n-4}\) (assuming CH ) for all n ≥ 4. All proofs are elementary and do not rely on theorems from set theory. Complete proofs can be found in [18].

Details

MFCS Conference 2009 Conference Paper

A Dynamic Algorithm for Reachability Games Played on Trees

Bakhadyr Khoussainov
Jiamou Liu
Imran Khaliq

Abstract Our goal is to start the investigation of dynamic algorithms for solving games that are played on finite graphs. The dynamic game determinacy problem calls for finding efficient algorithms that decide the winner of the game when the underlying graph undergoes repeated modifications. In this paper, we focus on turn-based reachability games. We provide an algorithm that solves the dynamic reachability game problem on trees. The amortized time complexity of our algorithm is O (log n ), where n is the number of nodes in the current graph.

Details