Erasmus+ gostovanje v Mehiki
Objavljeno: 09. jan. 2024
The Erasmus programme gave me the chance to visit the Center for Research and Advanced Studies of the National Polytechnic Institute in Mexico City, Mexico in December 2023. During my two-week stay, I interacted with leading researchers in the field of optimisation and attended several interesting tutorials. I also shared one of my recent research articles which analyzed the generalizability of algorithm selection models.
Dr. Urban Škvorc, Information and Communication Technologies
Through the Erasmus programme, I had the privilege of visiting the Center for Research and Advanced Studies at the National Polytechnic Institute in Mexico City in December 2023. This two-week trip allowed me to collaborate with top experts in optimisation, opening doors to new professional relationships and prospective partnerships.
During this visit, I showcased my recent work titled “How Far Out of Distribution Can We Go With ELA Features and Still Be Able to Rank Algorithms?”. This research delves into the resilience of Exploratory Landscape Analysis features for algorithm selection in varying single-objective optimization problem scenarios. In addition to my presentation, the trip encompassed various tutorials and social events, which were instrumental in enhancing network connections and laying the groundwork for future joint projects.
Gašper Petelin, Information and Communication Technologies
My PhD research is focused on representation learning for algorithm selection in the domain of single-objective optimization. Thanks to the Erasmus programme, I had the opportunity to visit the Center for Research and Advanced Studies of the National Polytechnic Institute in Mexico City, Mexico in December 2023. During my two-week stay, I was able to engage with leading researchers in the field of optimisation to establish new professional connections. I also presented one of my recent research articles titled: “TransOpt: Transformer-based Representation Learning for Optimization Problem Classification”, where we propose novel, transformer-based vectorized representations for single-objective optimization problems. The visit also included attending several tutorials and social activities to foster networking and establish future collaborations.
Gjorgjina Cenikj, Information and Communication Technologies
Nazaj na novice