Interview with Amar Halilovic: Explainable AI for robotics

In this meeting collection, we’re satisfying a few of the AAAI/SIGAI Doctoral Consortium individuals to learn even more regarding their study. The Doctoral Consortium gives a possibility for a team of PhD pupils to review and discover their study passions and job goals in an interdisciplinary workshop along with a panel of recognized scientists. In this newest meeting, we speak with Amar Halilovic, a PhD pupil at Ulm College.

Inform us a little bit regarding your PhD– where are you examining, and what is the subject of your study?

I’m presently a PhD pupil at Ulm College in Germany, where I concentrate on explainable AI for robotics. My study explores just how robotics can create descriptions of their activities in such a way that lines up with human choices and assumptions, especially in navigating jobs.

Could you offer us a review of the study you’ve executed until now throughout your PhD?

Up until now, I have actually created a structure for ecological descriptions of robotic activities and choices, particularly when points fail. I have actually discovered black-box and generative methods for the generation of textual and aesthetic descriptions. Additionally, I have actually been dealing with preparation of various description characteristics, such as timing, depiction, period, and so on. Recently, I have actually been dealing with techniques for dynamically choosing the very best description technique relying on the context and individual choices.

Exists a facet of your study that has been especially intriguing?

Yes, I locate it remarkable just how individuals translate robotic actions in different ways relying on the seriousness or failing context. It’s been particularly compensating to examine just how description assumptions change in various scenarios and just how we can customize description timing and web content appropriately.

What are your prepare for structure on your study until now throughout the PhD– what elements will you be examining following?

Following, I’ll be expanding the structure to include real-time adjustment, allowing robotics to pick up from individual responses and change their descriptions on the fly. I’m additionally preparing even more individual research studies to verify the performance of these descriptions in real-world human-robot communication setups.

Amar with his poster at the AAAI/SIGAI Doctoral Consortium at AAAI 2025.

What made you intend to examine AI, and, particularly, explainable robotic navigating?

I have actually constantly had an interest in the junction of people and devices. Throughout my research studies, I understood that making AI systems easy to understand isn’t simply a technological obstacle– it’s essential to count on and functionality. Robotic navigating struck me as a specifically engaging location due to the fact that choices are spatial and aesthetic, making descriptions both tough and impactful.

What recommendations would certainly you provide to a person thinking about doing a PhD in the area?

Select a subject that really delights you– you’ll be coping with it for numerous years! Additionally, construct an assistance network of advisors and peers. It’s simple to obtain shed in the technological job, yet cooperation and responses are crucial.

Could you inform us a fascinating (non-AI associated) reality regarding you?

I have actually lived and researched in 4 various nations.

Concerning Amar

Amar is a PhD pupil at the Institute of Expert System of Ulm College in Germany. His study concentrates on Explainable Expert system (XAI) in Human-Robot Communication (HRI), especially just how robotics can create context-sensitive descriptions for navigating choices. He integrates symbolic preparation and artificial intelligence to construct explainable robotic systems that adjust to human choices and various contexts. Prior to beginning his PhD, he researched Electric Design at the College of Sarajevo in Sarajevo, Bosnia and Herzegovina, and Computer Technology at Mälardalen College in Västerås, Sweden. Outdoors academic community, Amar delights in taking a trip, digital photography, and discovering links in between modern technology and culture.