Interview with Amina Mević: Machine learning applied to semiconductor manufacturing

In a collection of meetings, we’re satisfying a few of the AAAI/SIGAI Doctoral Consortium individuals to learn even more concerning their study. In this most recent meeting, we learn through Amina Mević that is using equipment discovering to semiconductor production. Figure out even more concerning her PhD study up until now, what makes this area so intriguing, and just how she discovered the AAAI Doctoral Consortium experience.

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

I am presently seeking my PhD at the College of Sarajevo, Professors of Electric Design, Division of Computer Technology and Informatics. My study is being accomplished in partnership with Infineon Technologies Austria as component of the Important Job of Usual European Passion (IPCEI) in Microelectronics. The subject of my study concentrates on establishing an explainable multi-output online assessment system based upon equipment discovering to anticipate the physical residential or commercial properties of steel layers in semiconductor production.

Could you offer us a summary of the study you’ve accomplished up until now throughout your PhD?

In the initial year of my PhD, I dealt with preprocessing complicated production information and preparing a durable multi-output forecast arrangement for online assessment. I teamed up with market professionals to recognize the procedure complexities and confirm the forecast versions. I used a projection-based choice formula (ProjSe), which straightened well with both domain name expertise and procedure physics.

In the 2nd year, I established an informative technique, developed to determine one of the most appropriate input functions for multi-output forecasts.

Exists an element of your study that has been specifically intriguing?

For me, one of the most intriguing facet is the harmony in between physics, maths, advanced innovation, psychology, and principles. I’m dealing with information gathered throughout a physical procedure– physical vapor deposition– utilizing ideas from geometry and algebra, specifically forecast drivers and their algebra, which have origins in quantum technicians, to boost both the efficiency and interpretability of artificial intelligence versions. Working together carefully with designers in the semiconductor market has actually additionally been mind-blowing, specifically seeing just how descriptions can straight sustain human decision-making in high-stakes settings. I really feel genuinely recognized to strengthen my expertise throughout these areas and to perform this multidisciplinary study.

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

I prepare to concentrate much more in a timely manner collection information and establish informative approaches for multivariate time collection versions. Furthermore, I plan to examine facets of accountable AI within the semiconductor market and make sure that the services recommended throughout my PhD line up with the concepts laid out in the EU AI Act.

Exactly How was the AAAI Doctoral Consortium, and the AAAI seminar experience as a whole?

Going To the AAAI Doctoral Consortium was a fantastic experience! It offered me the chance to provide my study and get beneficial responses from leading AI scientists. The networking facet was similarly satisfying– I had motivating discussions with fellow PhD pupils and advisors from worldwide. The major seminar itself was stimulating and varied, with advanced study provided throughout a lot of AI subfields. It certainly enhanced my inspiration and offered me originalities for the last stage of my PhD.

Amina offering 2 posters at AAAI 2025.

What made you intend to research AI?

After finishing in academic physics, I discovered that task chances– specifically in physics study– were rather restricted in my nation. I started searching for functions where I can use the mathematical expertise and analytical abilities I had actually established throughout my researches. At the time, information scientific research seemed a suitable and encouraging area. Nonetheless, I quickly recognized that I missed out on the deepness and function of basic study, which was usually doing not have in market functions. That encouraged me to go after a PhD in AI, intending to get a deep, fundamental understanding of the innovation– one that can be used meaningfully and utilized in solution of mankind.

What suggestions would certainly you provide to somebody thinking about doing a PhD in the area?

Keep interested and open up to gaining from various techniques– specifically math, stats, and domain name expertise. See to it your study has an objective that reverberates with you directly, as that interest will certainly aid bring you via obstacles. There will be minutes when you’ll seem like surrendering, however prior to making any type of choice, ask on your own: am I simply tired? Often, remainder is the service to much of our troubles. Lastly, locate advisors and areas to share concepts with and remain influenced.

Could you inform us a fascinating (non-AI associated) truth concerning you?

I’m a massive scientific research outreach fanatic! I consistently offer with the Organization for the Innovation of Scientific Research and Modern Technology in Bosnia, where we run workshops and occasions to motivate youngsters and secondary school pupils to check out STEM– specifically in underserved areas.

Concerning Amina

Amina Mević is a PhD prospect and training aide at the College of Sarajevo, Professors of Electric Design, Bosnia and Herzegovina. Her study is performed in partnership with Infineon Technologies Austria as component of the IPCEI in Microelectronics. She made a master’s level in academic physics and was granted 2 Golden Badges of the College of Sarajevo for attaining a grade point average more than 9.5/ 10 throughout both her bachelor’s and master’s researches. Amina proactively volunteers to advertise STEM education and learning amongst young people in Bosnia and Herzegovina and is committed to enhancing the study atmosphere in her nation.