This Week's Discoveries | 10 March 2020
- Tuesday 10 March 2020
- This Week's Discoveries
Niels Bohrweg 2
2333 CA Leiden
- De Sitterzaal
From stress to success: how actinobacteria exploit life without a cell wall
Dennis Claessen (IBL)
Dennis is an associate professor at the Institute of Biology and the current head of the cluster Microbial Sciences. With his lab, he studies morphogenesis and phenotypic heterogeneity in filamentous actinomycetes, which are prolific antibiotic producers. The Claessen lab pursues several research projects that span areas of multicellular growth and development, stress-adaptation and microbial evolution. By using multidisciplinary approaches, Dennis aims to tackle fundamental questions and to use the generated knowledge to improve the industrial exploitation of these organisms.
Most bacteria build a cell wall, which provides protection to cells in harmful conditions. However, some bacteria have the surprising ability to shed their wall in stressful situations. With a recently awarded Vici grant, I will investigate how bacteria do this and how they can actually profit from this wall-deficient state.
Make machine learning better for scientists across all domains
Jan van Rijn (LIACS)
Jan obtained his PhD in Computer Science in 2016 at Leiden University. During his PhD, he made several funded research visits to the University of Waikato (New Zealand, 3 times) and University of Porto. After obtaining his PhD, he worked as a post-doc the Machine Learning lab in Freiburg, headed by Prof. Dr. Frank Hutter, after which he moved to do a post-doc at Columbia University, in the City of New York.
His research aim is to democratize the access to Machine Learning tools across all entities in society, and his research interests include fundamental Computer Science, Automated Machine Learning and Data Science.
Data Science and Machine Learning are at the basis of many scientific discoveries across various Scientific domains. By allowing Machine Learning models, we are able to discover and research more complicated patterns than possible when
resorting to human expertise.
However, machine learning techniques are not easy to wield and require substantial training to be applied adequately and to have the results interpreted correctly. The field of Automated Machine Learning (AutoML) develops tools that can help domain scientists and experts in applying machine learning tools to their data. It is my vision that in order to bring this one step further, like in any other science, the field of Machine Learning should learn from prior experiments what paradigms and approaches are successful, and which are misleading.
During my PhD, we developed OpenML.org, an online experiment database with results from previous Machine Learning experiments. In this talk I will elaborate on the knowledge that we can gain from this, and how this can be applied to make machine learning better for scientists across all domains.