Organization of Challenges
This research axis is devoted to organizing challenges, typically to advance the domain of AutoML and AutoDL (coll. Google Zurich). Challenges (scientific competitions) are used as a means of stimulating research in machine learning, promoting reproducible science, and engaging a diverse community of engineers, researchers, and students to learn and contribute to advance the state-of-the-art. A huge diversity of industrial, academic and educational challenges (1,300 as of dec. 2022) have been supported within the public Codalab platform headed by Isabelle Guyon, that we actively use it in our own research and teaching. The data-centric and axiomatic perspectives raised by challenge processes and criteria are investigated in Zhengying Liu and Adrien Pavao’s PhDs.
… more to come
