AILA 2024

Keynote Speeches

Title: From AgentSpeak to Reinforcement Learning: the Journey of Developing Intelligent Autonomous Agents

Weiru Liu
Professor at the University of Bristol

Bio:Professor Weiru Liu holds Chair of Artificial Intelligence (AI) at the University of Bristol. Currently, she is a Co-Director on the EPSRC Centre for Doctoral Training in Future Autonomous and Robotic Systems: Towards Ubiquity (FARSCOPE-TU, 2019-2028). She is also a Co-Investigator on the ESRC funded Centre for Sociodigital Futures (2023-2027), where she leads research on examining social implications of AI and the importance of social science in designing future AI technologies. Prior to joining the University of Bristol in 2017, she held the Chair of AI at Queen’s University Belfast (QUB), UK. She obtained her PhD in 1995 from University of Edinburgh. Her research interests include, explainable AI, data-driven intelligent autonomous systems; planning; machine learning; cyber-physical systems; large-scale sensor network data analytics, and event modelling, reasoning and correlation in uncertain environments, with a wide range of applications such as security, healthcare, robotics.

Abstract:With the rapid advances of AI technologies, the prospect of developing and deploying intelligent autonomous systems (IAS) has never been so promising. However, there have been several key challenges when developing such systems, especially in uncertain and dynamic environments. Central to any IAS is an “agent” (such as a robot, or a software entity) which shall be able to perceive the world around it, to make sense of what is happening, to make decisions individually or collectively, and finally to respond by taking appropriate actions. Translating these abilities into technical terms, the agent shall be able to (i) observe the world (or the environment) either fully to partially (or not at all) through some sensors; (ii) analyse and combine multimodal information from different sources, coupled with its current (and/or previous) knowledge and beliefs to comprehend what is happening now; (iii) make decisions (under uncertainty) based on the modelling of and reasoning about the current situation; and (iv) decide the optimal action(s) to take in order to achieve its goal(s) or maximize its incentives. In this talk, I will mainly focus on topics (iv) and (iii) that are directly relevant to planning. I will examine the journey of AI planning, from its original logic-based forms, including Belief-Desire-Intention (BDI) agent, such as AgentSpeak, to online planning based on Markov Decision Process (MDP). I will then discuss how the progress in reinforcement learning (RL) could help to generate plans via policy learning. Finally, I will touch upon the most recent attempts of generating plans using LLMs, especially on the quality and validity of such plans. As a concluding remark, I will very briefly introduce and discuss topics in explainable AI, illustrating how black-box RL can be explained.

Title: Credible AI Models and Systems for Autonomous Decision Making

Jun Liu
Professor in Computer Science, Director of Artificial Intelligence Research Centre (AIRC) at School of Computing, Ulster University, Northern Ireland, UK.

Bio:Dr Jun Liu is currently a Professor in Computer Science, Director of Artificial Intelligence Research Centre (AIRC) at School of Computing, Ulster University, Northern Ireland, UK. Before he joined Ulster, he was a Postdoc at The University of Manchester, and a Postdoc at Belgian Nuclear Research Centre. He received BSc. and MSc. degrees in Applied Mathematics, and PhD. degree in Information Engineering from Southwest Jiaotong University, China, in 1993, 1996, and 1999, respectively. He worked in the field of AI for many years. His current research is focused on two themes: 1) trust and explainable data-knowledge integrated AI decision model/system with applications in management, engineering, and industry field etc. (e.g., safety and risk analysis; policy decision making; security/disaster management; anomaly detection and behavioural analysis for fin-crime; and heath care and smart home); 2) logic and automated reasoning methods for intelligent systems including software verification and automated theorem proving. In particular: resolution-based automated reasoning methods, algorithm and tools with applications (including software verification and automated theorem proving); lattice-valued logics with focus on handling incomparability, inconsistency and imprecision. He has authored or co-authored over 270 publications with over 6800 citations, the H-index 34 (ISI Web of Science); the H-index (Google Scholar) as 39 for all publications; and the i10-index (Google Scholar) as 109 for all publications. He has been awarded over £18 Million of research funding from various funding bodies as grant holder and principle co-investigator. He is the current Chair of IEEE CIS Emergent Technologies Technical Committee (ETTC), also the past Chair of IEEE System, Man and Cybernetics (SMC) Ireland Chapter (SMC28). He is an IEEE Senior Member including IEEESMC and IEEECI. He serves as an Associated Editor of IEEE Transaction on Fuzzy Systems, Knowledge-Based Systems, Human-Centric Computing and Information Sciences, and International Journal of Computational Intelligence Systems; also an Editor of Information Fusion, Journal of Universal Computer Science, and International Journal of Knowledge and Systems Science. He serves as chairs or co-chairs and program committees of a number of international conferences and workshops. He is a Fellow of the Higher Education Academy, and teaches at both undergraduate and postgraduate level.

Abstract:In today’s world, building trustworthy AI systems is paramount as AI becomes more prominent across the globe. Trustworthy AI represents the evolution of AI, and offers opportunities for industries to create AI system that are transparent, explainable, fair, robust, trust and reliable, especially in high-risk and safety-critical applications. This actually concerned two challenges: trustworthy AI model and trustworthy AI system, which are different problems and have to be handled in different ways, but they are closely related and both have to be achieved for the real applications, especially in safety-critical application. The talk aims to cover both aspects in a coherent but high level way: their motivation, the key ideas insight and the state-of-the-art, followed by our group’s research work illustrated with case studies in the explainable AI model and trustworthy system based on logic and automated reasoning.

Title: Speaking Logic in Computer Science

Zhiming Liu
Professor at Southwest University

Bio:Zhiming Liu is a professor at Southwest University. He obtained his master's degree from the Institute of Software, Chinese Academy of Sciences, in 1988, and his Ph.D. in Computer Science from the University of Warwick in 1991. From 1988 to 2015, he worked at the University of Warwick, the University of Leicester, and the United Nations University International Institute for Software Technology (UNU-IIST, Macau). He returned to China full-time in 2016 to teach at Southwest University, and from 2021 to 2022, he was a professor at Northwestern Polytechnical University. His main research areas theories and methods of software, focusing on trustworthy software methods and models and design of human-cyber-phyiscal systems. His notable academic contributions include fault-tolerant and real-time programming methods based on model transformations; probabilistic Duration Calculus for system reliability analysis; theory of semantics and refinement of object-oriented program; formal model-driven software development method rCOS; and modeling theories and methods for the software architecture of human-cyber-physical systems.

Abstract:Discuss the basic concepts, ideas and relations of formal logic, computational theory, and programming languages, showing the fundamental nature of mathematical logic to computer science and systems. We then briefly discuss the differences between deep learning and computation and programming based on deductive logic, and the scope of capabilities of deep learning.

Title: Intelligent Radio (IR) - A New Design Methodology for Wireless Communications and Spectrum Awareness

Tianfeng Yan
Professor at Lanzhou Jiaotong University

Bio:Tianfeng Yan is a professor in the School of Electronic and Information Engineering of Lanzhou Jiaotong University, and the director of the Institute of Digital Signal Processing and Software Radio of Lanzhou Jiaotong University. He has presided over one Gansu Provincial Scientific and Technological Progress Award, seven provincial and ministerial scientific and technological progress awards, and one People's Liberation Army Scientific and Technological Progress Award, and was selected as Gansu Provincial Leading Talent in 2009 and Gansu Provincial Top Leading Talent in 2020. His main research interests include radio monitoring, digital signal processing, software radio and related fields.

Abstract:Intelligent Radio (IR) is a new wireless communication technology based on the physical layer where optimized AI models can be incorporated in the transmitter and receiver side. For a communication system, the AI models in the transmitter and receiver can be self-corrected by online training based on the system's operating channel, which results in better channel adaptation for that communication system. For a spectrum sensing system, the effectiveness of sensing can also be improved.

Title: Generalized Closure System

Jun Zhang
Professor at Department of Psychology and Department of Statistics University of Michigan, Ann Arbor

Bio:Jun Zhang is with the University of Michigan as a Full Professor of Psychology and of Statistics. He is currently on leave there to take up the position of Professor at the newly-established Shanghai Institute for Mathematics and Interdisciplinary Sciences (SIMIS) for which he is a co-founder. He has held visiting positions at the University of Melbourne (Australia), CNRS Marseille (France), University of Waterloo (Canada), RIKEN Brain Science Institute (Japan), Center for Mathematical Sciences and Applications (CMSA) at Harvard University, and Shanghai Advanced Institute of Finance (China). He has served as the President, Vice President, and Executive Board Member of the Society for Mathematical Psychology, and as a member of the Council and a Member-at-Large on the Governing Board of the Federation of Associations in Brain and Behavioral Sciences (FABBS). He is a Fellow of the Association for Psychological Sciences (APS) and a Fellow of Psychonomic Society. He is the founding co-editor of the journal Information Geometry, has served as an Associate Editor of the Journal of Mathematical Psychology, and is a member of editorial boards of various journals. He directs the M3 Lab (“Mind, Machine and Mathematics”) conducting research in neuronal and brain signal analysis, computation vision, cognitive modeling, machine learning, brain-like computation and artificial intelligence. His current research effort is devoted to information geometry and geometrization of science of information.

Abstract:It is well-known that topological spaces can be axiomatically defined by the topological closure operator, i.e., Kuratowski Closure Axioms. Equivalently, they can be axiomatized by other set operators reflecting primitive notions of topology, such as interior operator, boundary operator, or derived-set operator (or dually, co-derived-set operators, exterior operators). Together, they provide the semantics of topological space. This talk is about weakening topological closure within a generalized closure system. It will be shown that the above six operators can all be weakened appropriately such that their relationships remain intact as in topological systems. And, giving an arbitrary subset, all points can be classified into equivalent classes (belong to its closure, boundary, exterior, etc), such that topological semantics can be fully relaxed to generalized closure systems. Since knowledge/learning space is an accessible set system (anti-matroid), with its own closure operator obtained from supplying the generalized closure operator with one additional axiom (related to anti-exchangeability), this result can have implications for providing semantics to the theory of knowledge/learning spaces. (Joint work with Dr. Yinbin Lei.)

	Full registeration as a non-member 20 July, 2024	Full registeration as a member 20 July, 2024	Registration as student 20 July, 2024
Price	2,200.00(CNY)	1,870.00(CNY)	1,100.00(CNY)

单位名称:	中国人工智能学会
开户银行:	中国工商银行北京新街口支行
账号:	0200002909200166203
行号:	102100000290
转账请务必备注:	AILA2024+姓名+论文ID号

Honorary Chairs

Ruqian Lu,
Academy of Mathematics and Systems Science,Chinese Academy of Sciences

Conference Chairs

Yang Jun,
Lanzhou Jiaotong University,China

Guo-Qiang Zhang,
University of Texas Houston,

Program Chairs

Luis Soares Barbosa,
Minho University,Portugal

Songmao Zhang,
Academy of Mathematics and Systems Science, Chinese Academy of Sciences, China

Organization Chairs

Kaijun Wu,
Lanzhou Jiaotong University,
China

Publicity Chairs

Yan Lu,
Lanzhou Jiaotong University,
China

Publication Chairs

Yiming Tang,
Hefei University of Technology,
China

Program Committee Members

Michal Baczynski, Faculty of Science and Technology, University of Silesia in Katowice, Katowice 40-007, Poland

Cungen Cao, Institute of Computing Technology, Chinese Academy of Sciences, China

Shaowei Cai, State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences

Yanhui Chen, College of Communication, Xi'an University of Electronic Technology, China

Shifei Ding, College of Computer Science, China University of Mining and Technology

Daqing Deng, Guangzhou City Institute of Technology, China

Jian Gao, College of Information Science and Technology, Northeast Normal University, China

Lluis Godo, PhD， Artificial Intelligence Research Institute, Bellaterra, Spain

Leandro Gomes, University of Lille, France

Xiaolong Jin, Institute of Computing Technology, Chinese Academy of Sciences, China

Fengkui Ju, Department of Philosophy, Beijing Normal University, China

Yong Lai, Jilin University, China

Ang Li, Changchun Institute of Optical Precision Instruments and Physics, Chinese Academy of Sciences

Fanzhang Li, Suzhou University, China

Jian Li, Jilin Agriculture University, China

Qin Li, College of Software Engineering, East China Normal University, China

Weizhuo Li, Nanjing University of Posts and Telecommunications, China

Yangyang Li, Academy of Mathematics and Systems Science, Chinese Academy of Sciences, China

Zhanshan Li, College of Computer Science and Technology, Jilin University, China

Beishui Liao, Institute of Logic and Cognition, Zhejiang University, China

Huawen Liu, Shandong University, China

Lin Liu, Tsinghua University, China

Renren Liu, Xiangtan University, China

Weiru Liu, University of Bristol, UK

Jun Liu, School of Computing, Faculty of Computing, Engineering and the Built Environment, Ulster University, UK

Alexandre Madeira, University of Aveiro, Portugal

Wenji Mao, Institute of Automation, Chinese Academy of Sciences, China

Luis Martinez, Computer Science Department, University of Jaén, Spain

Dantong Ouyang, College of Computer Science and Technology, Jilin University, China

Haiyu Pan, College of Computer and Information Security, Guilin University of Electronic Technology, China

Jihong Pei, Shenzhen University, China

Meikang Qiu, Texas A&M University Commerce, USA

Rosa Mª Rodríguez Domínguez, Computer Science Department, University of Jaén, Spain

Yun Shang, Academy of Mathematics and Systems Science, Chinese Academy of Sciences, China

Yanhong She, Xi'an Shiyou University, China
Joerg Siekmann, German Research Center for Artificial Intelligence (DFKI), Germany

Liang Sun, College of Software, Dalian University of Technology, China

Xianyong Tang, Sichuan University, Law College, China

Yiming Tang, Hefei University of Technology, China

Hongwei Tao, Zhengzhou University of Light Industry, China

Constantine Tsinakis, Department of Mathematics, Vanderbilt University, Nashville, TN, USA

Hui Wang, School of Electronics, Electrical Engineering and Computer Science, Queen's University Belfast, UK

Zhu Wang, Sichuan University Law College

Bin Wei, Zhejiang University, China

Hengyang Wu, School of Computer and Information Engineering, Shanghai Polytechnic University, China

Maonian Wu, Huzhou University, China

Zhongdong Wu, Lanzhou Jiaotong University, China

Juanying Xie, Shaanxi Normal University, China

Yun Xie, Institute of Logic and Cognition, Sun Yat sen University, China

Minghui Xiong, Zhejiang University, China

Youjun Xu, College of Computer Science and Information Technology, Daqing Normal University, China

Yuxin Ye, College of Computer Science and Technology, Jilin University, China

Changsheng Zhang, College of Software, Northeast University, China

Guangjun Zhang, Southwest University of Political Science and Law, China

Jian Zhang, State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences

Min Zhang, East China Normal University, China

Qinghua Zhang, Chongqing University of Posts and Telecommunications, China

Yong Zhang, Shenzhen University, China

Yuanrui Zhang, Southwest University, China

Bin Zhao, Shaanxi Normal University, China

Jian Zhao, College of Computer Science and Technology, Changchun University, China

Yang Zhao, Shenzhen University, China

Hongjun Zhou, Shaanxi Normal University, China

Li Zou, Shandong Jianzhu University, China