I am a postdoctoral researcher at UC Berkeley interested in algorithmic aspects of machine learning and artificial intelligence, with a focus on scalability, effciency and social impact. I am holding an SNF Early Postdoc.Mobility fellowship and I am hosted by Prof. Moritz Hardt. Prior to that I worked as a research scientist at IBM Research Zurich and contributed to the development of the Snap ML library. I have obtained my PhD from ETH Zurich where I was affiliated with the Data Analytics Laboratory and supervised by Prof. Thomas Hofmann.
When training machine learning models in production we often care about speed: fast training time allows short development cycles, offers fast time-to-insight, and after all, it saves valuable resources.
The goal of System-aware algorithm design is to achieve fast training through efficient utilization of compute resources available in modern heterogeneous systems.
We demonstrate [NeurIPS'18] that this approach can lead to several orders of magnitude reduction in training time compared to standard system-agnostic methods.
To achieve this we incorporate knowledge about system-level bottlenecks into the algorithm design. In particular, we develop new principled tools and methods for training machine learning models focusing on: compute parallelism [NeurIPS'19][ICML'20], hierarchical memory structures [HiPC'19][NeurIPS'17], accelerator units [FGCS'17] and interconnect bandwidth in distributed systems [ICML'18]. Most innovations of this research have been integrated in the IBM Snap ML library and help diverse companies improve speed, efficiency and scalability of their machine learning workloads.
Whenever we use supervised learning in social settings, we almost never make predictions for predictions' sake, but rather to inform decision making within some broader context.
Hence, our choice of predictive model can lead to changes in the way the broader system behaves.
We call such predictions performative.
We introduce the framework of performative prediction to supervised learning [ICML'20] and address challenges faced in stochastic optimization when the deployment of a model triggers performative effects in the data distribution it is being trained on [NeurIPS'20]. As a subfield of learning theory, performative prediction is only starting to receive attention from the community and there are many exciting open challenges to explore.