My name is Madalina (Ina) Fiterau, I'm a fifth year PhD student in Machine Learning at Carnegie Mellon University, working with Artur Dubrawski of the Auton Lab.
My main interests are query-specific models, anomalous pattern detection and active learning with dimensionality reduction. My current research explores the idea of building small ensembles of low-dimensional components which are applicable to significant subsets of data, such that every sample can be handled using one of these submodels or using a sparse mixture of them. The assignment of samples to sub-models and the dimensionality reduction for the learners on the submodels are performed jointly, using a convex formulation.
Application-wise, I am working on analysing vital sign data obtained by monitoring patients admitted to the Intensive Care Unit. My interest in the use of machine learning for medical applications. This is the focus of the MLCDA@NIPS workshop I co-organized with Jenna Wiens and Finale Doshi-Velez.
I've started off as a software engineer, having received my BEng from Politehnica Timisoara, Romania. Coding is something I enjoy, which is one of the reasons why I've twice interned at Google.
- Artifact adjudication for health status alerts. This is a collaborative project with a team from the University of Pittsburgh. We work with a cardio-respiratory monitoring system designed to process multiple vital signs indicative of the health status of ICU patients. The system issues an alert whenever some form of instability requires attention, but in practice, a substantial fraction of these alerts are artifacts. A subset of the data has been manually labeled and the aim is to use that subset to determine which of of the unlabeled samples are worth the experts' attention.
[active learning; dimensionality reduction; ensemble methods]
- Detection of nuclear threats from vehicles scanned at the US border.
One of the many projects under way at the Auton Lab, this presents a multiclass discrimination problem where expert feedback is available.
The models need to be interpretable. Another interesting aspect of this dataset is that the training and test samples come from different distributions.
[multilabel classification; interpretable models; transductive learning]
- Finding explanations in datasets by exploring areas in low-dimensional projections of the feature space.
I've implemented some heuristic approaches as well as LPs which explicitly minimize entropy-based objective functions.
[feature selection; information theory; linear programming]
- Regression for Informative Projection Retrieval (RIPR) - a regression-based algorithm which identifies informative projections by optimizing over a matrix of point-wise loss estimators, learning compact models for classification, clustering, and regression.
[ensemble methods; feature selection; projection retrieval; query-specific models]
- Vital Sign Monitoring System that predicts clinical alarms and detects anomalous patterns in patient vital signs.
This project uses the waveform and treatment data in the Mimic II dataset.
[dynamic Bayesian Networks; variational methods]
- Finding sets of high-accuracy regions in low-dimensional subspaces. The framework uses any given hypothesis class and enables visualization of the classification process.
[ensemble methods; query-specific models]
- Fiterau M, Dubrawski A, A Unified View of Informative Projection Retrieval, ICMLA 2013
- Fiterau M, Dubrawski A, Chen L, Hravnak M, Clermont G, Pinsky MR, Automatic identification of artifacts in monitoring critically ill patients, ESICM-LIVES 2013
- Fiterau M, Dubrawski A, Detecting Artifacts in Clinical Data through Projection Retrieval, ICML-HEALTH Workshop 2013
- Madalina Fiterau, Artur Dubrawski, Projection Retrieval for Classification, NIPS 2012
- Madalina Fiterau, Artur Dubrawski, Trade-offs in Explanatory Model Learning, Data Analysis Project
- Madalina Fiterau, Artur Dubrawski, Can Ye, Real-time Adaptive Monitoring of Vital Signs for Clinical Alarm Preemption, international Society for Disease Surveillance Annual Conference 2011
- Madalina Fiterau, Olga Ormond, Gabriel-Miro Muntean, Performance of Handover for Multiple Users in Heterogeneous Wireless Networks, IEEE Conference on Local Computer Networks 2009
- Undergraduate thesis: Handover Algorithm Design and Simulation, coordinators Dr. Gabriel Muntean, Prof. Ioan Jurca
10th of December 2013
The Machine Learning for Clinical Data Analysis Workshop took place at NIPS.
Many thanks to our speakers, panelists, poster presenters and co-organizers - Jenna Wiens and Finale Doshi!
4th of December 2013
The RIPR algorithm presented at ICMLA 2013.
view camera ready and slides.
30th of September 2013
The paper Automatic Identification of Alarm Artifacts in Monitoring Critically Ill Patients presented at ESICM-LIVES.
view electronic poster.
20th of June 2013
RIPR models for artifact adjudication shown at the ICML-HEALTH Workshop.
view camera ready and poster.
3rd of December 2012
The RECIP algorithm was shown at NIPS 2012.
view camera ready and poster.
3rd of December 2012
'Feature-Task Co-clustering Regression' was presented at WiML 2012.
view abstract and poster.
21st of March 2012
Data Analysis Project. Committee: Prof. Artur Dubrawski, Prof. Geoff Gordon, Prof. Jeff Schneider
view slides pptx|pdf.
12th of December 2011
the algorithm `Explanation-Oriented Partitioning' was presented at the WiML workshop in Granada.
7th of December 2011
the paper `Real-time Adaptive Monitoring of Vital Signs for Clinical Alarm Preemption' presented at the iSDS Annual Conference in Atlanta.
29th of November 2011
Aiming for the Moon, an outline of the Lander visual registering project was presented during the ML Journal Club.