"Large Language Model and Multi-Agent System Reading Club" starts this Saturday!

Introduction

"Large Language Model and Multi-agent System Reading Club" will be shared for the first time this Saturday night at 20:00. This time, Guo Taicheng, a doctoral student in computer science at the University of Notre Dame, and Li Guohao, the founder of the currently popular multi-agent framework CAMEL and a postdoctoral fellow at the University of Oxford, will give lectures! More paper authors from Tsinghua University, Peking University, Zhejiang University, MIT, UIUC and other universities will appear in turn, so stay tuned!

Sharing content outline

Guo Taicheng: Overview of LLM multi-agent

LLM multi-agent background introduction

Basic elements of multi-agent

Interface

Profiling

Communication

Capabilities

Current multi-agent research work types

Overview

Framework

Agents for Problem-Solving

Agents for World Simulation

Orchestration and Capabilities Acquisition

Commonly used data sets for multi-agent research

Challenges and research opportunities from different perspectives

Li Guohao: CAMEL-Exploring the multi-agent society of large language model LLM

Basic concepts

Mind exploration

Communicative Agents

Building a multi-agent system

Role-playing framework

Inception prompt

instruction-following

Experiment Display and current challenges

Generating scenarios

Experimental results

Current challenges

Speaker introduction

##Guo Taicheng: A doctoral candidate in computer science at the University of Notre Dame, whose supervisor is Professor Xiangliang Zhang.

Research direction: Use LLM and LLM-based Agents to solve reasoning and planning problems in scientific fields, AI4Science and recommendation systems. I have published articles at conferences such as NeurIPS and WWW, and have won KDD Cup, NeurIPS BlackBox Optimization Challenge, IEEE Computer Society Challenge and other machine learning competition champion/runner-up.

#Li Guohao: Founder of CAMEL, postdoctoral fellow at Oxford University, artificial intelligence researcher, initiator of CAMEL-AI and DeepGCNs open source projects. Committed to building intelligent agents that can perceive, learn to communicate, reason and act. He is also a core member of PyG.org. He received his PhD in Computer Science from King Abdullah University of Science and Technology under Professor Bernard Ghanem. During his PhD, he worked as a research intern at Intel ISL. He visited ETHz CVL as a visiting researcher and has worked at Kumo AI.

Research directions: including autonomous agents, graphical machine learning, computer vision and embedded artificial intelligence. He has published relevant papers in top conferences and journals such as ICCV, CVPR, ICML, NeurIPS, RSS, 3DV, and TPAMI.

Host Introduction

Zhao Shiyu, Distinguished Researcher at the School of Engineering of West Lake University, Director of the Intelligent Unmanned System Laboratory, and winner of the National Overseas High-Level Talent Youth Project By. He graduated from Beijing University of Aeronautics and Astronautics with a bachelor's degree and a Ph.D. from the National University of Singapore. Before joining Westlake University, he was a lecturer in the Department of Automatic Control and Systems Engineering at the University of Sheffield, UK. His research focuses on issues such as control decision-making in multi-robot systems, and is committed to studying next-generation robotic systems that are interesting, useful, and challenging. In addition, the English textbook "Mathematical Foundations of Reinforcement Learning" written by him will be jointly published by Tsinghua University Press and Springer Press. It has currently received 1.7K stars on GitHub, and the course video has been viewed more than 700,000 times online.

Ren Qinyuan received his PhD degree in control science and engineering from Zhejiang University in 2008. Then he joined the Department of Computer and Electrical Engineering of the National University of Singapore to engage in control research on underwater unmanned intelligent systems. Starting from July 2015, he joined the Information Research Institute of Singapore's Agency for Science, Technology and Research as a robotics research scientist. Since 2017, he has joined Zhejiang University as a professor. Dr. Ren Qinyuan is currently an IEEE Senior Member and serves as the editor of several international academic journals. He is mainly engaged in research work on motion control, bionic intelligence and human-computer interaction.

Dr. Cui Jinqiang is a senior engineer at Pengcheng Laboratory. His research areas include multi-agent systems, simultaneous mapping and localization (SLAM), and high-precision 3D reconstruction. Dr. Cui received his PhD from the National University of Singapore and previously completed his master's and bachelor's studies at Northwestern Polytechnical University. He has won multiple honors in international micro-UAV competitions and published many academic papers in related fields.

Large Language Model and Multi-agent System Reading Club

We have invited many cutting-edge scholars to share their views, including Guo Taicheng, Li Guohao, Qian Chen, Wang Zhenhailong, Xu Yuzhuang, Yang Zonghan , Liu Zijun, Hongxin Zhang, Zhang Jintian, Dong Yihong, Liang Tian, ​​Yilun Du, etc., more topics are still being recruited and supplemented. If you are interested in large models and multi-agent systems, welcome to join us and share. You can also come and make friends, and the most important thing is to study together!

Mainly related references

[ 1 ] Guo T, Chen X, Wang Y, et al. Large language model based multi-agents: A survey of progress and challenges [ J ] . arXiv preprint arXiv:2402.01680, 2024.

[ 2 ] Sumers T R, Yao S, Narasimhan K, et al. Cognitive architectures for language agents [ J ] . arXiv preprint arXiv:2309.02427, 2023.

[ 3 ] Hong S, Zheng X, Chen J, et al. Metagpt: Meta programming for multi-agent collaborative framework [ J ] . arXiv preprint arXiv:2308.00352, 2023.

[ 4 ] Li G, Hammoud H A A K, Itani H, et al. Camel: Communicative agents for "mind" exploration of large scale language model society [ J ] . arXiv preprint arXiv:2303.17760, 2023.

直播information

Time:

20:00-22:00 on the evening of March 2, 2024 (this Saturday).

How to participate:

Scan the QR code to participate in the multi-agent reading club, join the group chat, get access to review the series of reading clubs, and participate in front-line scientific research with the community Workers communicate with business practitioners to jointly promote the development of the cutting-edge field of multi-agent.

Special thanks to the unit

Datawhale

An open source organization focusing on the field of AI, bringing together many outstanding learners, with a mission for the learner, and Learners grow together.

Agent42

An ecological open platform dedicated to promoting the integration and innovation of industry, academia and research on AI Agents

The above is the detailed content of "Large Language Model and Multi-Agent System Reading Club" starts this Saturday!. For more information, please follow other related articles on the PHP Chinese website!