I study robots and their brains at EPFL in Switzerland.
I was an intern
on the LHCb experiment at CERN, and am currently completing my
master thesis at the SNAP laboratory in Stanford. I'm trying to
optimize my research interests, so feel free to help update my gradients
by getting in touch! You can also check out my CV here.
These publications span several concurrent domains, from the application of Machine Learning in High Energy Physics (HEP) to the development of novel characterizations of media bias. This is the product of diverse research interests and a variety of research opportunities that I have been lucky enough to pursue.
News entities have to select and filter the coverage they broadcast through their respective channels, since the set of world events is too large to be treated exhaustively. The subjective nature of this filtering induces biases due to, among other things, resource constraints, editorial guidelines, ideological affinities, or even the fragmented nature of the information at a journalist's disposal. The magnitude and direction of this bias are, however, widely unknown. The absence of ground truth, the sheer size of the event space, or the lack of an exhaustive set of absolute features to measure makes it difficult to observe the bias directly, to characterize the leaning's nature and to factor it out to ensure a neutral coverage of the news.
In this work, we introduce a methodology to capture the latent structure of media’s decision process at a large scale. Our contribution is multi-fold. First, we show media coverage to be predictable using personalization techniques, and evaluate our approach on a large set of events collected from the GDELT database. We then show that a personalized and parametrized approach not only exhibits higher accuracy in coverage prediction, but also provides an interpretable representation of the selection bias. Last, we propose a method able to select a set of sources by leveraging the latent representation. These selected sources provide a more diverse and egalitarian coverage, all while retaining the most actively covered events.
In order to achieve the data rates proposed for the future Run 3 upgrade of the LHCb detector, new processing models must be developed to deal with the increased throughput. For this reason we aim to investigate the feasibility of purely data-driven 'holistic' methods, with the constraint of introducing minimal computational overhead, hence using only raw detector information. These filters should be unbiased - having a neutral effect with respect to the studied physics channels. In particular, the use of machine learning based methods seems particularly suitable, potentially providing a natural formulation for heuristic-free, unbiased filters whose objective would be to optimize between throughput and bandwidth.
Presenting a machine learning based approach to estimate the position and uncertainty of a VELO track’s closest to beam state. The resolution of this prediction as well as its ability to estimate the uncertainty is shown to be superior. While this is not a production ready solution yet, these preliminary results are promising and indicate that a machine learning based approach might provide an alternative to the simplified Kalman Filter.