Adrian Barbu
Professor, Department of Statistics, Florida
State University.
Currently Serving as Graduate Student Director.
Please note that I am not in the admissions committee, so I cannot answer any admissions related questions.
Research
interests:
- Machine Learning
- Computer Vision
- Medical Imaging
Current Research:
- Object Segmentation - Using PCA as a shape model for object segmentation.
(link)
- Large Scale Unsupervised Learning - Algorithms with theoretical guarantees for estimating mixtures of
Gaussians in the presence of outliers.(arxiv)
- Large Scale Supervised/Semi-Supervised Learning - Incremental methods for learning models from millions of
observations with thousands of classes.(IEEE)
- Online Learning - Updating sufficient statistics in an online fashion to be able to extract models of
different complexities at any time.(arxiv)
- Robust Learning - Learning models that know when the data is outside the training distribution. (arxiv)
- Feature Selection with Annealing - A generic method for feature selection and model learning
that outperforms boosting and methods based on sparsity inducing penalties such as L1, SCAD, and MCP. (arxiv)
Representative Publications
- C. Long, S. Nag, A. Barbu. PCA-UNet for Object Segmentation, IEEE International Conference on Image Processing (ICIP), 2024.
(link,
GitHub)
- Y. Zhou, K. Gallivan, A. Barbu. Scalable Clustering: Large Scale Unsupervised Learning of Gaussian Mixture Models with Outliers.
Journal of Computational and Graphical Statistics, 1-17, 2024.
(link,
arxiv,
GitHub)
- L. Sun, M. Wang, A. Barbu. A Novel Framework for Online Supervised Learning with Feature Selection. Journal of Nonparametric Statistics, 2024.
(link, arxiv, slides)
- H. Pabuccu, A. Barbu. Feature Selection for Forecasting. Financial Innovation 10, No. 27, 2024.
(link, arxiv)
- K. Han, A. Barbu. Large-Scale Few-Shot Classification with Semi-supervised Hierarchical k-Probabilistic PCAs. IJCNN, 2024.
(link)
- S. Liu, A. Barbu. Unsupervised Learning of Mixture Models with a Uniform Background Component.
(arxiv)
- N. Lay, A.P. Harrison, S. Schreiber, G. Dawer, A. Barbu. Random Hinge Forest for Differentiable Learning.
(arxiv)
- B. Wang, A. Barbu. Hierarchical Classification for Large-Scale Learning.
Electronics 12, No. 22, 4646, 2023.
(link,
GitHub)
- S. Manna, M. Wang, A. Barbu, C. V. Ciobanu. Machine-learning of Piezoelectric Coefficients for Wurtzite Crystals.
Materials and Manufacturing Processes, 38, No. 16, 2081-2092, 2023.
(link)
- Y.She, J.Shen, A. Barbu. Slow Kill for Big Data Learning. IEEE Trans. Information Theory, 69, No. 9, 5936-5955, 2023.
(IEEE)
- A. Barbu. Training a Two Layer ReLU Network Analytically. Sensors 23, No. 8, 4072, 2023
(arxiv,
link,
GitHub)
- M. Wang, A. Barbu. Online Feature Screening for Data Streams with Concept Drift.
IEEE Trans. on Knowledge and Data Engineering, 2023.
(arxiv,
IEEE,
GitHub)
- B. Wang, A. Barbu. Scalable Learning with Incremental Probabilistic PCA.
IEEE International Conference on Big Data, Special Session on Machine Learning for Big Data, 2022.
(pdf,
GitHub)
- O. Akal, A. Barbu. Fast 3D Liver Segmentation Using a Trained Deep Chan-Vese Model.
Electronics 11 No. 20, 3323, 2022
(link,
GitHub)
- C. Long, A. Barbu. A Study of Shape Modeling Against Noise. IEEE International Conference on Image Processing (ICIP), 611-615, 2022.
(pdf, slides)
- A. Barbu, H. Mou. The Compact Support Neural Network. Sensors 21 No. 24, 8494, 2021.
(arxiv,link)
- H. Huang, A. Barbu. Predicting Lane Change Decision Making with Compact Support. IEEE Intelligent Vehicles Symposium, 2021.
(pdf)
- Y.Guo, A. Barbu. A study of local optima for learning feature interactions using neural networks. IJCNN 2021
(arxiv)
- Y.Guo, Y. She, A. Barbu. Training Efficient Network Architecture and Weights via Direct Sparsity Control. IJCNN 2021
(arxiv)
- B. R. Bartoldson, A. S. Morcos, A. Barbu, G. Erlebacher. The Generalization-Stability Tradeoff in Neural Network Pruning.
Neural Information Processing Systems (NeurIPS), 2020,(arxiv)
- G. Dawer, Y.Guo, A. Barbu. Generating Compact Tree Ensembles via Annealing. IJCNN 2020 (arxiv)
- G. Dawer, Y.Guo, S. Liu, A. Barbu. Neural Rule Ensembles: Encoding Sparse Feature Interactions into Neural Networks.
IJCNN 2020 (arxiv)
- D. Li, A. Barbu. Training a Steerable CNN for Guidewire Detection. CVPR 2020,
(pdf)
- A. Barbu, S.C. Zhu. Monte Carlo Methods. Springer 2020
(Springer, Amazon)
- H. Huang, A. Barbu. Playing Atari Ball Games with Hierarchical Reinforcement Learning.
(arxiv)
- M. Wang, A. Barbu. Are screening methods useful in feature selection? An empirical study. PLoS One 14, No. 9 (2019)
(arxiv,link)
- O. Akal, A. Barbu. Learning Chan-Vese. ICIP 2019, Taipei, Taiwan
(pdf)
- D. Li, A. Barbu. Training a CNN for Guidewire Detection. ICIP 2019, Taipei, Taiwan
(pdf)
- N. Lay, Y. Tsehay, R. Cheng, S. Gaur, A. Barbu, L. Lu, B. Turkbey, P. Choyke , P. Pinto, R. Summers.
A Decomposable Model for Prostate Cancer Detection in Multi-Parametric MRI. MICCAI, 2018, Granada, Spain
(pdf)
- H. Mou, A. Barbu. Accurate Dictionary Learning with Direct Sparsity Control. ICIP 2018, Athens, Greece
(pdf,
GitHub)
- J. Anaya, A. Barbu. RENOIR - A Dataset for Real Low-Light Image Noise Reduction. Journal of Visual Comm. and Image Rep. 51, No. 2, 144-154, 2018 (arxiv,
link, data)
- A. Gupta, A. Barbu. Parameterized Principal Component Analysis. Pattern Recognition 78, No. 6, 215–227, 2018
(arxiv, link)
- A. Barbu, N. Lay, G.Gramajo. Face Detection with a 3D Model. "Academic Press Library in Signal Processing Volume 6: Image and Video Processing and Analysis and Computer Vision".
pp 237-259, 2018. Editors: R. Chellappa and S. Theodoridis.(arxiv, link)
- A. Barbu. A Directed Graph Approach to Active Contours. ICIP 2017 (pdf)
- A. Barbu, Y. She, L. Ding, G. Gramajo. Feature Selection with Annealing for Computer Vision and Big Data Learning. IEEE Trans. PAMI, 39, No. 2, 272-286, 2017.
(arxiv,
link,
GitHub)
- A. Barbu, L. Lu, H. Roth, A. Seff, R.M. Summers. An Analysis of Robust Cost Functions for CNN in Computer-Aided Diagnosis. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization 2016.
(pdf, link)
- L. Ding, A. Barbu. Scalable Subspace Clustering with Application to Motion Segmentation. Current Trends in Bayesian Methodology with Applications, Chapman & Hall/CRC Press.
Editors: Dipak K. Dey, Umesh Singh and A. Loganathan. pp 267-286, 2015 (pdf)
- A. Barbu, T.F. Wu, Y. N. Wu. Learning Mixtures of Bernoulli Templates by Two-Round EM with
Performance Guarantee. Electronic Journal of Statistics 8, No. 2, 3004-3030, 2014 (pdf, arxiv)
- A. Barbu, M. Pavlovskaia, S.C. Zhu. Rates for Inductive Learning of Compositional Models. AAAI Workshop Replearn 2013
(pdf)
- A. Barbu. Multi-Path Marginal Space Learning for Object Detection.
Academic Press Library in Signal Processing: Volume 4: Image, Video Processing and Analysis,
Hardware, Audio, Acoustic and Speech Processing. pp 271-291, 2013
(pdf)
- A. Barbu. Hierarchical Object Parsing from Structured Noisy Point Clouds. IEEE PAMI, 35, No. 7, 1649-1659, 2013.
(pdf, arxiv, slides)
- L. Ding, A. Barbu, A. Meyer-Baese. Learning a Quality-Based Ranking for Feature Point Trajectories. ACCV 2012
(pdf)
- L. Ding, A. Barbu, A. Meyer-Baese. Motion Segmentation by Velocity Clustering with Estimation of Subspace Dimension. ACCV Workshop DTCE 2012
(pdf)
- A. Barbu, N. Lay. An Introduction to Artificial Prediction Markets for Classification.
Journal of Machine Learning Research, 13, 2177-2204, 2012.
(pdf)
- A. Barbu, M. Suehling, X. Xu, D. Liu, S. K. Zhou, D. Comaniciu. Automatic Detection and Segmentation of Lymph Nodes from CT Data.
IEEE Trans Medical Imaging, 31, No. 2, 240-250, 2012.(pdf)
- W. Wu, T. Chen , A. Barbu, P. Wang, N. Strobel, S. Zhou, D. Comaniciu.
Learning-based Hypothesis Fusion for Robust Catheter Tracking in 2D X-ray Fluoroscopy.
CVPR 2011 (pdf)
- F. Bunea and
A.Barbu. Dimension reduction and variable selection in case control
studies via
regularized likelihood optimization. Electronic Journal of Statistics, 3,
2009. (pdf)
- A. Barbu. Training an Active Random Field for Real-Time
Image Denoising. IEEE Trans. Image Processing, 18,
November 2009. (pdf,
ppt)
- S. Seifert, A.
Barbu, S. Zhou, D. Liu, J. Feulner, M. Huber, M. Suehling, A.
Cavallaro, D. Comaniciu.
Hierarchical parsing and semantic navigation of full body CT data. SPIE
Medical Imaging, 2009 (pdf)
- Y. Zheng,
A. Barbu, B. Georgescu, M. Scheuering and D. Comaniciu.
Four-Chamber Heart Modeling and Automatic Segmentation for 3D Cardiac CT Volumes Using Marginal Space
Learning and Steerable Features. IEEE Trans Medical Imaging, November
2008. (pdf)
- L.
Lu, A. Barbu, M. Wolf, J. Liang, M. Salganicoff,
D. Comaniciu. Accurate Polyp Segmentation for 3D
CT Colonography Using Multi-Staged Probabilistic Binary Learning and
Compositional Model. CVPR 2008.(pdf)
- A. Barbu, V.
Athitsos, B. Georgescu, S. Boehm,
P. Durlak, D. Comaniciu. Hierarchical Learning of Curves:
Application to Guidewire Localization in Fluoroscopy. CVPR 2007 (pdf)
- A. Barbu, S.C. Zhu. Generalizing Swendsen-Wang
for Image Analysis. J. Comp. Graph. Stat. 16, No 4, 2007 (pdf)
- A. Barbu, S.C. Zhu. Generalizing Swendsen-Wang
to sampling arbitrary posterior probabilities, PAMI, 27,
August 2005 (pdf)
- A. Barbu,
S.C. Zhu. Multigrid and Multi-level
Swendsen-Wang Cuts for Hierarchic Graph Partition, CVPR 2004 (pdf)
Awards:
- Graduate Faculty Mentor Award, Florida State University, 2023
- 2011 Thomas A. Edison Patent Award. System and Method for Segmenting Chambers of a Heart in a Three Dimensional Image.
Patent No. 7,916,919. video
Funding:
- Fundamental Limits of Learning, DARPA, $142,000, 09/14/2016-12/31/2017. (PI on Subcontract from UCLA).
- Learning Homogeneous Knowledge Representation from Heterogeneous Data for Quantitative and Qualitative Reasoning in Autonomy.
DARPA, $42,000, 06/29/2015- 07/28/2016. (PI on Subcontract from UCLA)
- A Novel Platform for Biological Information Integration and Knowledge Discovery, NSF, $70,000. 07/01/2014 - 06/30/2015, 2015 (Co-PI)
- SEE on a Unified Foundation for Representation, Inference and Learning, DARPA, $257,000. (PI on subcontract from UCLA)
- MCS: Research on Detection and Classification of 2D and 3D Shapes in Cluttered Point Clouds. NSF, $400,000. (CO-PI)
- Statistical and Semantic Approaches for Object, Activity and Intent Recognition. ONR, $443,000 (CO-PI)
- Landmark Detection Using Discriminative Anatomical Network And Active Random Fields. Siemens, $31,000 (PI)
- Cooperative Systems: Task Allocation for Heterogeneous Agent Teams Via Stochastic Clustering Auctions. ARO, $16,000 (subcontract)
Invited Talks
- Object Segmentation: A Journey from Level Sets to Shape Denoising.
Keynote presentation at the International Conference in Medical Imaging and Computer-Aided Diagnosis (MICAD), 2022
(pdf)
- Learning Nonlinear Feature Interactions in the Data Starved Regime. 2021 Florida ASA Chapter Meeting, April 2021
(pdf)
- A Novel Framework for Online Supervised Learning with Feature Selection. 2021 JMM/AMS Meeting, January 2021
(pdf)
- Online Learning with Model Selection. NC State Univ., February 1st, 2019 (pdf)
- Artificial Intelligence and the Future of Humanity, CWT Solutions Group Meeting, Miami Beach, FL, February 2016.
(pdf)
- Face Detection with a 3D Model. UCLA, June 19, 2015
- Feature Selection with Annealing for Big Data Learning. Johns Hopkins University, October 28, 2014
- Feature Selection with Annealing for Regression and Classification. Temple University, October 15, 2014
- The Artificial Prediction Market, ICML Workshop on Markets, Mechanisms and Multi-Agent Models,
Edinburgh, July 1st, 2012 (pdf)
- Hierarchical Object Parsing from Noisy Point Clouds, Siemens Corporate Research, August 16th, 2011
- Artificial Prediction Markets for Classification, Regression and Density Estimation, UCLA, August 11th, 2011
- Automatic Detection and Segmentation of Lymph Nodes. NIH, December 8th 2010
- Supervised Aggregation of Classifiers using Artificial
Prediction Markets, SRCOS 2010 (pdf)
- Supervised Aggregation using Artificial Prediction Markets.
UCLA, November 10th, 2009
- Marginal Space Learning for Fast Object Detection in Medical Imaging.
Tutorial on Discriminative Learning Methods in Medical Imaging, MICCAI 2009
- Training an Active Random Field for Real-Time Image
Denoising. Max Plank Institute, Saarbrucken, Germany, July 16th, 2008
- The Swendsen-Wang Cuts Algorithm with Applications in
Computer Vision, Georgia Tech University, June 2008
- Active Random Fields for Real-Time Image Denoising, Siemens
Corporate Research, May 2008
- Hierarchical Image-Motion Segmentation using Swendsen-Wang Cuts,
Third Cape Cod MCMC Workshop
, Harvard, 2007
(ppt)
- A General Clustering Sampling Method for Bayesian Inference, Joint Statistical Meetings,
Minneapolis, August 10, 2005
- Swendsen-Wang for Perceptual Grouping.
Second Cape Cod Workshop on Monte Carlo Methods, 2004
Patents:
- J. Anaya, A. Barbu. System and Method for Image Processing using Automatically Estimated Tuning Parameters.
Patent no. 10,032,256
- J. Anaya, A. Barbu. System and Method for Generating a Dataset for Real Noise Reduction Evaluation.
Patent no. 9,591,240
- Y. Zheng, A. Barbu, B. Georgescu, M. Lynch, M. Scheuering, D. Comaniciu.
Method and system for generating a four-chamber heart model. Patent no 9,275,190
- A. Barbu. Systems and Methods for Training an Active Random Field for Real-Time Image Denoising.
Patent no 8,866,936
- A. Barbu, W. Zhang, N. Strobel, A. Galant, U. Bill, D. Comaniciu.
Method and system for catheter detection and tracking in a fluoroscopic image sequence. Patent no 8,396,533
- A. Barbu, M. Suehling, X. Xu, D. Liu, S.K. Zhou, D. Comaniciu.
Method and System for Automatic Detection and Segmentation of Axillary Lymph Nodes. Patent no 8,391,579
- W. Zhang, Y. Zhu, A. Barbu, R. Socher, S. Boehm, P. Durlak, D. Comaniciu.
Methods and Apparatus for Virtual Coronary Mapping. Patent 8,355,550
- L. Lu, A. Barbu, M. Wolf, S. Lakare, L. Bogoni, M. Salganicoff, D. Comaniciu.
Method and System for Polyp Segmentation for 3D Computed Tomography Colonography.
Patent 8,184,888
- L. Lu, A. Barbu, M. Wolf, S. Lakare, L. Bogoni, M. Salganicoff, D. Comaniciu.
User Interface for Polyp Annotation, Segmentation and Measurement in 3D Computed Tomography Colonography.
Patent 8,126,244
- R. Socher, A. Barbu, B. Georgescu, W. Zhang, P. Durlak, S. Bohm, D. Comaniciu.
Method and system for vessel segmentation in fluoroscopic images. Patent 8,121,367
- Y. Zhu, W. Zhang, A. Barbu, S. Prummer, M. Ostermeier, C. Reddy, D. Comaniciu.
System and Method for Coronary Digital Subtraction Angiography. Patent 8,094,903
- W. Zhang, A. Barbu, S. Prummer, M. Ostermeier and D. Comaniciu.
Method and System for Evaluating Image Segmentation Based on Visibility. Patent 8,086,006
- A. Barbu, L. Lu, L. Bogoni, M. Salganicoff, D. Comaniciu.
Method and System for Detection and Registration of 3D Objects Using Incremental Parameter Learning.
Patent 8,068,654
- A. Barbu, V. Athitsos, B. Georgescu, P. Durlak, D. Comaniciu.
System and Method for Online Optimization of Guidewire Visibility in Fluoroscopic Systems.
Patent 8,050,482
- A. Barbu, Y. Zheng, Y. Zhing, B. Georgescu, D. Comaniciu.
System and Method for Detecting an Object in a High Dimensional Space.
Patent 8,009,900
- B. Georgescu, P. Durlak, V. Athitsos, A. Barbu, D. Comaniciu.
System and Method for Simultaneously Subsampling Fluoroscopic Images and Enhancing Guidewire Visibility.
Patent 7,970,191
- W. Zhang, A. Barbu, S. Prummer, M. Ostermeier, C. Reddy, D. Comaniciu.
System and Method for Coronary Digital Subtraction Angiography. Patent 7,940,971
- Y. Zheng, A. Barbu, B. Georgescu, M. Scheuering, D. Comaniciu.
System and Method for Segmenting Chambers of a Heart in a Three Dimensional Image.
Patent 7,916,919
- S.K. Zhou, J. Shao, J. Dowdall, A. Barbu, B. Georgescu, D. Comaniciu.
System and Method for Using a Similarity Function to Perform Appearance Matching in Image Pairs.
Patent 7,831,074
- A. Barbu, V. Athitsos, B. Georgescu, P. Durlak, S. Boehm,
D. Comaniciu. System and Method for Detecting and Tracking
a Guidewire In A Fluoroscopic Image Sequence. Patent 7,792,342
- A. Barbu, L. Lu, M. Wolf, L. Bogoni, M. Salganicoff. Interface Utilisateur pour une
Annotation, une Segmentation et une Mesure de Polype en Colonoscopie Virtuelle 3D.
Europe Patent WO/2009/042034
- A. Barbu, L. Bogoni, D. Comaniciu. System and Method For
Detecting A Three Dimensional Flexible Tube In An Object. Patent 7,783,097
- Z. Tu, A. Barbu. Probabilistic
Boosting Tree Framework for Learning Discriminative Models. Patent 7,702,596
- Z. Tu, X. Zhou, D. Comaniciu, L. Bogoni, A. Barbu. System
and Method for Using Learned Discriminative Models to Segment Three dimensional Colon Image Data.
Patent 7,583,831
- Z. Tu, A. Barbu, D. Comaniciu. Method for Detecting Polyps in a Three Dimensional Image Volume. Patent 7,558,413
Teaching:
- Applied Machine Learning - STA 4634/5635 (syllabus) Spring 2008-2016, Fall 2016-present, Spring 2022-present
- Statistical Computing with Python - STA 5934-0002 (syllabus) Spring 2019-2021
- Applied Linear Regression - STA 4203/5207 (syllabus) Fall 2007-2013,2015-2016
- ANOVA and Design of Experiments - STA 4202/5206 (syllabus) Spring 2016-2017
- Introduction to Applied Statistics - STA 5126 (syllabus) Fall 2012
- Medical Image Analysis, (syllabus) Spring 2010,2012
Education:
2000-2005: Ph.D. Computer
Science, University of California,
Los Angeles
1995-2000: Ph.D. Mathematics,
Ohio
State University
1990-1995: B.Sc. Mathematics,
University of Bucharest, Romania
If you want to
reach me, my address
is:
Adrian Barbu
Department of Statistics
Florida State University
Tallahassee, FL 32306
Phone:(850) 290-5202
E-mail: 1abarbu23@46fsu.edu67 (remove all
the numbers)