Our group hosts computer scientists, geoscientists, a mathematician, a physicist. We work with many interdisciplinary collaborators.
Our main funded projects currently focus on global scale analysis of biospheric change, evolutionary processes, reconstructing past climates and environmental change. In data science our primary focus is on methods for handling concept drift, and reasoning with evolving data. We are also concerned how to make robust, transparent and interpretable models in general.
Data science for understanding evolutionary processes and how life works in general using data of the global fossil record, present day animal communities, climate, vegetation and environemnt models and sensory observations. Research directions include analysis of macroevolutionary processes, large scale reconstruction of paleoclimate, understanding evolution of faunal communities and contexts of human evolution, and providing quantitative insights for mitigating climate change.
Keywords: Red Queen, ecometrics, macroevolution, biospheric change and conservation palaeobiology, human environments
Our long tern focus in computational research has been on handling concept drift in machine learning. Traditional machine learning and data mining methods rely on the assumption that data distribution stays the same during model training and operation. As the world is continuously changing, so evolve data that describe it. We develop machine learning methods that can diagnose themselves and adapt to changing data distribution over time.
Keywords: concept drift, change detection, evolving streaming data
Applied data science is about translating human expertise from various domains to computational proxies and then modeling human expertise in a data-driven way. We have worked in data science in various domains, including: palaeobiology, forestry, biomedicine, chemical engineering, hardware design, traffic and mobility analysis, finance, process mining, sales analysis and customer profiling.
Towards hardware-driven design of low-energy algorithms for data analysis by Žliobaitė et al. 2014 in SIGMOD Record.
Regression models tolerant to massively missing data: a case study in solar radiation nowcasting by Žliobaitė et al. 2014 in Atmospheric Measurement Techniques.
Online Detection of Shutdown Periods in Chemical Plants: A Case Study by Martin-Salvador et al. 2014 in Procedia Computer Science.
Predicting respiratory motion for real-time tumour tracking in radiotherapy by Krilavičius et al. 2016 in IEEE CBMS.
Environmental control of growth variation in a boreal Scots pine stand – a data-driven approach by Kulmala et al. 2016 in Silva Fennica.
Herbivore teeth predict climatic limits in Kenyan ecosystems by Žliobaitė et al. 2016 in PNAS.
Fairness-aware machine learning studies in which circumstances algorithms may become biased towards or against groups of people, and how to make predictive models free from such biases, when data, on which they are built, may be biased, incomplete, or even contain past discriminatory decisions.