Tutorial 1 Session, Day 3: 20th, 15:00-17:00JST
Title of the tutorial: Automated Negotiation in Supply Chain Management Competition
Acronym of the tutorial: SCM
The name and email address of the presenter(s): Yasser Mohammad
The name and email address of the presenter(s):
Principal Researcher, Data Science Research Laboratories, NEC, Japan Visiting Researcher, AIST, RIKEN, Japan
Associate Professor, Assiut University, Egypt firstname.lastname@example.org
Keywords: Automated Negotiation, Machine Learning, NegMAS, Supply Chain Management, ANAC
Duration: 2 hours.
Automated negotiation between intelligent agents is attracting more attention from the research community especially with the wider market penetration of intelligent agents and the need to coordinate their behavior. The International Automated Negotiating Agents Competition (ANAC) provided stimulation for this research since its intro- duction in 2010. Since 2019, a new league was added to ANAC focusing on application of automated negotiation in a realistic business-like Supply Chain Management scenario (SCML). This tutorial will introduce the audience to the SCML and walk them through the development of an agent for the competition highlighting the research challenges involved.
Tutorial Web site: http://yasserm.com/prima-2020-scml-tutorial/
Tentative Detailed outline of the tutorial
The tutorial will consist of two main parts with a 10-min break.
Theoretical Session (40min) This part of the tutorial introduces the ANAC competition and the SCML describing the canonical structure of agent decomposition for the competition
The negotiation problem (5min) Different definitions of the negotiation problem, negotiation protocol, main differences between negotiations and auctions.
ANAC (5min) A short history of the ANAC competition.
SCM World (10min) Introduces the game design for SCML.
Why SCML (5min) Provides the rationale behind the SCML design.
Agent decomposition (15min) Introduces the decomposition of SCML agents into two inter-related components: The trading strategy, negotiation strategy, production strategy, and signing strategy.
Development Environment (15min) A hands-on installation and configuration tutorial
Installing SCML (5min) Goes through the process of installation and configuration for the SCML package with the underlying NegMAS platform and the repository of SCML agents.
Running a simulation (5min) Goes through the process of running a single simulation and understanding log files.
Running tournaments (5min) Introduces tournaments and their parameters, as well as methods, ensure that com- parisons are fair when running tournaments.
Live Development of an agent (20min) Provides a hands-on demo for developing an agent for the SCM world.
An ML-based trading strategy (10min) Develops a trading strategy that uses ML for predicting various aspects of the SCM world simulation.
Putting it all together (10min) Combines the developed trading strategy with built-in components to create a complete agent for SCML
Development Example (30min) This final part of the tutorial aims at giving the participants confidence that they could grasp the general structure of an SCML agent and that they can develop an agent for the competition and/or participate in related research in the future. We propose two alternatives here. Depending on the readiness of the participants will either do a hands-on hackathon or a study of existing strategies for the league.
Examples of Agent Strategies (30min) This part of the tutorial will walk the participants through the strategies used by some of the finalists in SCML2020 highlighting the different ways for improving upon the builtin agents.
Hands-on hackathon (30min) Participants will be given 30min to develop their own agent focusing on modifying a single component of the newly developed agent. The main goal of this step is not to come up with a strong agent but to make sure that the participant has understood the structure of the agent and build confidence in her/his ability to develop a real agent for future competitions.
Conclusions (5min) The tutorial will be wrapped-up by a summary of the information introduced in the first session about automatic negotiation and will provide interesting directions of research inviting the audience to actively participating in pushing forward this exciting domain.
The presenter is a Principal Researcher at NEC (Data Science Research Laboratories), Japan and an Associate Professor at Assiut University, Egypt. He also holds visiting researcher positions at AIST and RIKEN national research institutes, Japan. He received his Ph.D. from Kyoto University, Japan in 2009 in the area of intelligence science and technology. Recipient of four best paper awards from ICCAS 2012, IEEE/SICE SII 2011, IEA/AIE 2009, and IEA/AIE 2009. Author of Conversational Informatics: A data-intensive approach (Springer, 2015) and Data mining for Social Robotics (Springer, 2016). His research focuses on automatic negotiation, multi-agent systems, and applications of data mining techniques to time-series data. Published over 100 international publications in these areas with an h-index of 16. Co-founder of Ayonix Inc., Japan, and Tebex IT, Egypt, and a senior IEEE member.
Tutorial 2 Session, Day 3: 20th, 17:00-19:00JST
Title of the tutorial: Negotiation Support Systems
Acronym of the tutorial: NSS
The name and email address of the presenter(s):
Catholijn M. Jonker email@example.com;
Interactive Intelligence Group, Delft University of Technology &
LIACS Leiden University
Reyhan Aydoğan firstname.lastname@example.org
Computer Science, Ozyegin University &
Interactive Intelligence Group, Delft University of Technology
Keywords: automated negotiation, negotiation support, intelligent agents, machine learning, knowledge technology, multi-criteria optimization, human-agent experiment design.
Duration: 2 hours.
Negotiation support systems (NSS) provide decision support in problems involving multiple decision makers on a multitude of issues (multi-actor multi-criteria optimization). While some people are very good at negotiation, others have difficulty in reaching optimal outcomes and mostly end up with suboptimal outcomes. Improving on negotiation outcomes can be done by training people before they enter the negotiation, by delegating the negotiation to others, or by supporting them during the negotiation. How to develop Intelligent support for all of these options is a rich and vibrant research area.
The negotiation support systems tutorial will motivate and introduce novices to the rich research field of automated negotiation and negotiation support agents that are part of the PRIMA research community.
The attendees will receive an overview of existing practices and methodologies for negotiation support systems and negotiating agents. In particular, we discuss systems
That train human negotiators
That support humans during a negotiation
That advice human negotiators on what to bid and when to stop
Furthermore, we will discuss how we can measure the effectiveness of mechanisms specifically designed for supporting people during the bidding phase of negotiations. In particular, we will examine the features of existing negotiation support systems.
The attendees will experiment with the system during the tutorial, and we discuss the effectiveness of the system as studied in an experiment setup according to the tutorial’s teachings. We round off the tutorial with a discussion of the grand research challenges for automated negotiation and negotiation support systems. We introduce them to the international annual Automated Negotiating Agents Competition in which the industry participates and is increasingly interested.
We will upload the tutorial material (e.g., slides, software, etc.) online before the event.
Tentative Detailed outline of the tutorial
17.00 - 17.10 Introduction
17.10 - 17.20 Overview of existing systems
17.20 - 17.40 Mechanisms for bidding support
17.40 - 17.50 Experimental design for evaluation decision support systems
17.50 - 17.59 break
18.00 - 18.10 Introduction to the experiment
18.10 - 18.20 Practice with a negotiation support system on a water management scenario
18.20 - 18.40 Experiment
18.40 - 18.50 Experience exchange and evaluation of the experimental results
18.50 - 18.59 Grand research challenges for automated negotiation and negotiation support
18.59 - 19.00 closing
Prof. dr. Catholijn Jonker is head of the Interactive Intelligence group of the faculty of Electrical Engineering, Mathematics, and Computer Science, TU Delft. Jonker is also full professor of Explainable Artificial Intelligence at the Leiden Institute of Advanced Computer Science of Leiden University. She is president of IFAAMAS, board member and a Fellow of EurAI, member of the Academia Europaea, president of ICT Platform of the Netherlands, member of the Royal Holland Society of Sciences and Humanities, member of the CLAIRE National Advisory Board for The Netherlands. In the past she was chair of the Dutch Network of Female Full Professors and of De Jonge Akademie of the Royal Netherlands Academy of Arts and Sciences. Prestigious grants are the NWO VICI (1.5 M€, 2007) personal grant negotiation support systems, and NWO Gravitation consortium grants on “Hybrid Intelligence” (19 M€ subsidy, 41.9 M€, 2019) of which she is vice-coordinator, and “Ethics of Socially Disruptive Technologies” (18 M€ subsidy, 23.6 M€, 2019) of which she is co-applicant.
Dr. Reyhan Aydoğan is an assistant professor in the Department of Computer Science at Özyeğin University and guest researcher in the Interactive Intelligence Group at TU Delft. She received her PhD. degree in 2011 in Computer Engineering from Boğaziçi University, Istanbul, Turkey and after which she joined the Interactive Intelligence Group at Delft University of Technology as a postdoctoral researcher. As a guest researcher, she visited the Center of Collective Intelligence at MIT in 2013; Intelligence Systems Group at Norwegian University of Science and Technology in 2015 and Nagoya Institute of Technology in 2017. She is co-organizer of the International Automated Negotiating Agent Competition since 2014 and co-organizers of several workshops among which ACAN and COREDEMA. She is serving as a program committee member in reputable AI conferences such as AAAI, AAMAS, IJCAI, PRIMA and ECAI. Her research focuses on the modeling, development and analysis of intelligent agents that integrate different aspects of intelligence such as reasoning, decision making and learning. She is well-known for her research on qualitative preference modeling, automated negotiating agents and negotiation protocols. Besides autonomous agents, she also designs and develops decision support systems in particular negotiation support systems. Her aim is to support human decision makers in complex and dynamic environments, which also requires the design of effective human computer interaction (e.g. preference elicitation). Her career project is about human-robot negotiation.