HUMAINE Research Project

Project Overview

HUMAINE aims to advance the state-of-the-art in human-computer interaction by researching a) methods that can capture human-centered subjective notions such as behavior, experience and creativity more reliably and b) representations that can be learned with less human generated data and more efficiently.

HUMAINE focuses on the computational and AI systems that can best understand the human element, by conducting innovative research on a) methods that can capture human-centered subjective notions such as behavior, experience and creativity more reliably and b) representations that can be learned with less human generated data and more efficiently.

HUMAINE is undertaken by researchers in the Institute of Digital Games at the University of Malta. It has received funding by the Malta Council for Science and Technology (MCST) under the FUSION R&I: Research Excellence Programme (Project number: REP-2022-017).

Annotation Interfaces

HUMAINE extends current research by the Institute of Digital Games on capturing subjective notions such as behavior, experience and creativity. To achieve this, a set of interaction paradigms and interfaces will be designed, developed and tested in order to capture the most by asking a human for their subjective experience with the least amount of bias and cognitive load.

Ordinal Modeling

HUMAINE expands the current focus of machine learning (ML) research to capturing an objective truth towards subjectively-defined tasks that require human input such as human behavior and human emotion recognition. HUMAINE fuses ordinal representations with deep preference learning to learn more from less data and train ML models effectively from fewer human demonstrations.

Outputs

Konstantinos Makantasis, Kosmas Pinitas, Antonios Liapis and Georgios N. Yannakakis: "From the Lab to the Wild: Affect Modeling via Privileged Information," in IEEE Transactions on Affective Computing, 2023 (accepted).
Matthew Barthet, Chintan Trivedi, Kosmas Pinitas, Emmanouil Xylakis, Konstantinos Makantasis, Antonios Liapis and Georgios N. Yannakakis: "Knowing Your Annotator: Rapidly Testing the Reliability of Affect Annotation," in Proceedings of the 11th International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW).
Tech demo of the new chromatic rank annotation interface (developed in collaboration with David Melhart)
Open-source AI library for Learning Under Priviledged Information for affect signals (deveoped in collaboration with Kosmas Pinitas).

Research Team

Dr. Antonios Liapis: Senior Lecturer at the Institute of Digital Games.

Prof. Georgios N. Yannakakis: Professor and Director of the Institute of Digital Games and a Co-Founder and Research Director (Malta) of modl.ai.

Dr. Iro Voulgari: Postdoctoral fellow at the Institute of Digital Games.

Mr. Konstantinos Sfikas: PhD student under the supervision of Dr. Liapis and Prof. Yannakakis.