Human and Machine Intelligence in Collaborative Decision Making (HMI)

IEEE SSCI 2020 will run as a virtual conference.

Machine learning (ML) and artificial intelligence (AI) have experienced a surge in recent years. They have been used to develop models for facilitating decision making processes in different areas up to date. The development of these models is based on the idea that computers can process big data and make predictions whilst it is often hard for human experts. However, humans are more skilled to work with unstructured information and deal with uncommon situations. It is expected that developing decision making systems where humans and machines collaborate with each other can provide improvements for decision making. Ideally, decisions should not be made by individuals, decision making can often be improved by involving multiple information sources. These systems would be useful in several areas such as clinical decision-making, Defence commanding, autonomous driving, criminal punishment prediction, etc. In some of these areas, machine intelligence has not been introduced much e.g. judiciary systems. In a judiciary system, decisions are usually made by human which is the society’s expectation. In this system, the goal of collaborative decision making is to develop an informed decision making platform to make an unbiased and fair decisions for punishment. In some other areas, AI models have been applied largely e.g. clinical decision making. However, the reliability of these models is always under question. A collaborative decision making can provide a trust between clinicians and AI models. In some areas like Defence, decision making requires the synchronisation and fusion of sensors and intelligence feeds as well as the integration of human input at different points in the various networks in a timely manner. Although some studies have addressed the importance of these collaborative decision making recently, there are few works on developing these types of decision making.

There are several benefits and challenges which are involved with collaborative decision making. Interpretability and transparency are the issues with which machine intelligence models face. Many situations, such as clinical decision making, are critical that should be understood by human, allowing for intervention if the models lead to undesired results. A proper collaborative decision making not only can improve the accuracy and reliability of decision making, but also can be more interpretable. Another important issue that should be considered is uncertainty. It is usually difficult for decision makers to be certain when examining options. This stems from the nature of human thinking and the complication of the options. Decisions and analytical outcomes from AI and ML model are also associated with uncertainty due to prevalence of uncertainty in data, which comes from many reasons such as imprecise measurement systems, natural variation and linguistic expression. These uncertainties should be considered in collaborative decision making. In addition, collaborative decision making provides and opportunity to utilise uncertainty as a source of information to minimise information loss.

This special session aims to showcase the importance of collaboration between human and machines and initiate developing models on collaborative decision making with applications in different areas. The special session will attract researcher and practitioners who work on decision making, explainable AI, uncertainty and human computer interaction (HCI). Special attention will be devoted to group decision making in a collaborative environment, handling uncertainty in prediction and pattern recognition for incomplete and imbalance data.

Topics

Collaborative decision making between human and machine
AI interpretability improvement using human collaboration
Handling uncertainty in machine learning using expert decision making techniques
Group decision making between human experts and machine intelligence
Human and AI collaboration in clinical decision making
Collaborative decision making in judiciary systems
Machine learning using uncertainty quantification
Information loss mitigation in machine learning
Ensemble learning under uncertainty and fuzzy logic
Human intelligence in evolutionary computation
Semi-supervised learning in a collaborative environment