Madalina Fiterau: “Extracting Interpretable Models from Structured and Multiresolution Timeseries Data”

Madalina Fiterau will present her work on January 27th at 11AM.

It will be online at https://cnrs.zoom.us/j/92127397588?pwd=UTUzQXRmbDBSQ2hLbEpDSDYrcnVrQT09

 

Title

Extracting Interpretable Models from Structured and Multiresolution Timeseries Data

Abstract

The prevalence of smartphones and wearable devices and the widespread use of electronic health records have led to a surge in biomedical time series data that is noisy and non-uniform, collected at an unprecedented scale. Models analyzing these time series, however, are often not transparent, which reduces their trustworthiness and impedes their eventual deployment. This talk presents a series of methods that bridge the gap between performance and interpretability in time series modeling. In the first part of the talk, I will focus on techniques for featurizing vital signs time series and leveraging low-dimensional structure to enable visualization and annotation by domain experts. I will show how our framework, VIPR, was used to adjudicate false alerts in vital signs collected from patients in a step-down unit, with minimal effort on the part of the clinicians. In the second part of the talk, I will present two deep learning models: ShortFuse, which performs temporal representation learning by fusing time series and structured information and MultiWave, which uses wavelet decomposition and a gating mechanism to augment any deep learning time series model with components operating at the intrinsic frequencies of the signals. I’ll present our state of the art results on the prediction of surgical outcomes in children with cerebral palsy, forecasting the progression of osteoarthritis from subjects’ physical activity as well as the WESAD and COVID-19 benchmarks.

Bio

Ina Fiterau completed a PhD in Machine Learning at Carnegie Mellon University in Fall 2015, where she was a member of the Auton Lab. Between Fall 2015 and Fall 2018, she was a Postdoctoral Fellow in the Mobilize Center at Stanford University. She joined the College of Information and Computer Sciences at UMass in September 2018. Ina is the recipient of the Marr Prize for Best Paper at ICCV 2015 and of Star Research Award at the Annual Congress of the Society of Critical Care Medicine 2016. She was also the recipient of the Manning IALS Research Award in 2019.

Comments are closed.