Machine learning is being used more and more in different tasks all around the society. Commonly machine learning replaces humans in tasks, which used to require monotonous manual monitoring, but also completely new kinds of solutions and services are being born. To find the optimal machine learning methods for each application and to develop novel algorithms – besides proper understanding of various machine learning methods – it is necessary to closely collaborate with the experts on the application domain to solve real problems in a way that will benefit end users most. My research on machine learning has allowed me to have interesting collaboration, for example, with boat manufacturers to produce algorithms for autonomous boats, with historians and photojournalists to extract information from the World War II photographs, and with biologists to develop automated biomonitoring.
I received my PhD in signal processing at Tampere University of Technology in 2017. In September 2019, I started working as a senior research scientist at the Finnish Environment Center after receiving Postdoctoral Researcher funding from the Academy of Finland for 2019-2022 for a project “Advanced machine learning methods for biomonitoring.” In the project, I will develop novel algorithms for automatic image-based taxa identification focusing on detecting rare and previously unseen taxa, solving problems related to varying imaging conditions, and hierarchical classification. I also plan to work on combining image-based classification with DNA metabarcoding.