Publications

ViDi: Descriptive Visual Data Clustering as Radiologist Assistant in COVID-19 Streamline Diagnostic

Published in preprint, 2021

Sahithya Ravi, Samaneh Khoshrou and Mykola Pechenizkiy

In this paper, we investigate state-of-the-art models for covid detection from chest-xrays. We generate homogeneous clusters in terms of disease severity and interpret the clusters using favorable and unfavorable saliency maps, which visualize the class discriminating regions of an image. These human-interpretable maps complement radiologist knowledge to investigate the whole batch at once. Keywords: Interpretable Machine learning, human-in-the-loop

Download here

Openml-python: an extensible python api for openml

Published in Journal of Machine Learning Research (JMLR), 2021

Matthias Feurer, Jan N. van Rijn, Arlind Kadra, Pieter Gijsbers, Neeratyoy Mallik, Sahithya Ravi, Andreas Müller, Joaquin Vanschoren and Frank Hutter. My contributions can be found here.

The paper describes the OpenML Python package which is the python framwork of the popular OpenML.org platform and used extensively by many studies in the field of AutoML and meta-learning. Keywords: Machine learning, Open data science, AutoML

Download here

Evaluation, modeling and optimization of coverage enhancement methods of NB-IoT

Published in 2019 IEEE 30th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC), 2019

Sahithya Ravi, Pouria Zand , Mohieddine El Soussi and Majid Nabi

In this paper, we formulate a constrained convex optimization problem in the domain of Internet of things, specifically - Narrow Band IoT. The optimization strategy aims to allocate different coverage enhancement features in such a way that the latency is minimum and the reliability is maintained. We attempt to solve this problem analytically and verify our solution using network simulations. Keywords: Constrained optimization, Lagrange optimization, Internet of Things

Download here