science (6051) machine learning (3353) aws (1306) deep learning (1107) automl (16)
My goal is to build automated learning systems that are able to configure themselves without any human intervention. More specifically, my research focus is on hyperparameter and architecture search of deep neural networks as well as more general AutoML.
I am an applied scientist in the long-term science team of AWS SageMaker, where I work on large-scale distributed hyperparameter and architecture search. Prior to that, I was a PhD student at the University of Freiburg under the supervision of Frank Hutter . In 2015, me and my collaborators from the University of Freiburg won the ChaLearn AutoML Challenge. I co-organized the workshop on neural architecture search at ICRL 2020 and ICLR 2021, respectively. Together with Giovanni Zappella , Arber Zela and Jovi
Machine learning is one of the most disruptive technologies of our time. However, a recurring criticism of machine learning is that, in practice, training a model is often more of an art than a science.
Dive into Deep Learning — Dive into Deep Learning 1.0.3 documentation
Peter Richtarik
Dive into Deep Learning — Dive into Deep Learning 1.0.3 documentation
AutoML Seminar
Giovanni Zappella
Jovita Lukasik
Numair Mansur · Personal Page