For best experience please turn on javascript and use a modern browser!
You are using a browser that is no longer supported by Microsoft. Please upgrade your browser. The site may not present itself correctly if you continue browsing.

Master AI students publish conference papers

Floor Eijkelboom

Floor, a second-year student in the ELLIS honours programme of the MSc AI, published part of his master's thesis at the International Conference on Machine Learning (ICML). ICML is known as one of the few top-tier conferences in machine learning research. In this work, Floor combined two fundamentally different approaches to improving the expressivity of graph neural networks to leverage both benefits when doing e.g. molecular prediction tasks and modelling N-body systems. Floor was supervised in this project by Dr. Erik Bekkers and Rob Hesselink (AMLAB of the Informatics Institute).

Piyush Bagad

Piyush, also second-year student in the Master AI programme, published a paper at Computer Vision and Pattern Recognition (CVPR), the world’s most prestigious AI conference. According to Google Scholar, CVPR currently ranks as the fourth most cited scientific publication, just after Nature, The New England Journal of Medicine and Science. It is also the first time a Master AI student from UvA publishes at CVPR. Piyush has been enhancing video-language models with a sense of time, through efficient contrastive learning. He was advised by Prof. Cees Snoek at UvA and Dr Makarand Tapaswi at IIIT Hyderabad (India).

Konstantinos Papakostas and Irene Papadopoulou

Two other Master AI students, Konstantinos and Irene, have published a short paper at the highly regarded conference of the Association for Computational Linguistics. Working under the guidance of Professor Evangelos Kanoulas of the Information Retrieval Lab (IRLab), the students developed the project initially during the "Information Retrieval 2" elective course in their second year. They researched difficulties that language model can face when generating long-form answers that are both coherent and accurate, to make the models more reliably deployed in production.