Interactive Learning: Human in the Loop System Design for Active Human–Computer Interactions

Dear Colleagues,

The consideration of the active participation of persons in human–computer systems involves several considerations over and above what occurs in more passive man–machine interactions. Typically, this involves subjectivity in which humans use their insights to make judgments for the application solution. Interactive machine learning uses machine learning implementations that are trained, optimized, evaluated, and exploited through an intuitive human–computer interface. Having a human analyst in the loop is of high interest due to quality control, accountability, and complex subject matter expertise not otherwise readily automated. It should also be possible to use implicit human feedback to adjust a system by observing human problem-solving activity.

Potential Topics:

Cognitive load—can use models to adjust decisions.

How much should be pre-training (i.e., learning for the average user) versus how much should be interactive or personalized (i.e., for fine-tuning to a specific user)?

Response design and modalities of interaction—use of natural/implicit human feedback signals such as natural language, speech, eye movements, facial expressions, and gestures during the interaction.

Effective interactions—speed and number of interactions. Human preferences or an internal reward that is non-stationary and changes over time. Limitations can be due to a lack of trust, usability, and productivity, especially when adapting to unforeseen classes and changes in the mission context.

Specific systems architecture—AI and machine learning application to the problem; human trust issues for different architectures.

Case studies—for example, case studies such as image segmentation and region digitization in GIS are desirable.

Prof. Dr. Fred Petry
Dr. Chris J. Michael
Prof. Dr. Derek T. Anderson
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Information is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• human in the loop
• cognitive load
• interaction modalities
• cognitive feedback
• online machine learning
• active machine learning