@tue.nl
postdoctoral researcher, Mathematics and Computer Science
Eindhoven University of Technology (
My research interests span across sequential decision making and fairness in machine learning.
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Rosa Van Tuijn, Tianqin Lu, Emma Driesse, Koen Franken, Pratik Gajane, and Emilia Barakova
IOS Press
We introduce WeHeart, a personalized recommendation device that aims to gradually increase physical activity levels in cardiac rehabilitation. The importance of physical activity in cardiac rehabilitation as a means of reducing associated morbidity and mortality rates is well-established. However, forming physical activity habits is a challenge, and the approach varies depending on individual preferences. Our solution employs a Random Forest classification model that combines both measured and self-reported data to provide personalized recommendations. We also propose to make use of Explainable AI to improve transparency and foster trust.
Rosa van Tuijn, Tianqin Lu, Emma Driesse, Koen Franken, Pratik Gajane, and Emilia Barakova
Springer Nature Switzerland
Dennis Collaris, Pratik Gajane, Joost Jorritsma, Jarke J. van Wijk, and Mykola Pechenizkiy
Springer Nature Switzerland
AbstractLocal surrogate learning is a popular and successful method for machine learning explanation. It uses synthetic transfer data to approximate a complex reference model. The sampling technique used for this transfer data has a significant impact on the provided explanation, but remains relatively unexplored in literature. In this work, we explore alternative sampling techniques in pursuit of more faithful and robust explanations, and present LEMON: a sampling technique that samples directly from the desired distribution instead of reweighting samples as done in other explanation techniques (e.g., LIME). Next, we evaluate our technique in a synthetic and UCI dataset-based experiment, and show that our sampling technique yields more faithful explanations compared to current state-of-the-art explainers.
Danil Provodin, Pratik Gajane, Mykola Pechenizkiy, and Maurits Kaptein
IEEE
We consider a special case of bandit problems, named batched bandits, in which an agent observes batches of responses over a certain time period. Unlike previous work, we consider a more practically relevant batch-centric scenario of batch learning. That is to say, we provide a policy-agnostic regret analysis and demonstrate upper and lower bounds for the regret of a candidate policy. Our main theoretical results show that the impact of batch learning is a multiplicative factor of batch size relative to the regret of online behavior. Primarily, we study two settings of the stochastic linear bandits: bandits with finitely and infinitely many arms. While the regret bounds are the same for both settings, the former setting results hold under milder assumptions. Also, we provide a more robust result for the 2-armed bandit problem as an important insight. Finally, we demonstrate the consistency of theoretical results by conducting empirical experiments and reflect on optimal batch size choice.