Ilija Bogunovic

 

I am broadly interested in machine learning algorithms that are theoretically well-founded, robust and practically applicable. 

 

Contact Address 

EPFL STI IEL LIONS

ELD 243 (Batiment EL), Station 11

CH-1015, Lausanne

+41(0) 21 693 46 91

email: ilija.bogunovic@epfl.ch

Biography

Since 2014, I am a doctoral assistant at EPFL, supervised by Prof. Volkan Cevher and cosupervised by Prof. Jarvis Haupt. ​​​​​​I received my B.S. degree in electrical and computer engineering from the University of Novi Sad and M.S. degree in computer science from ETH Zurich.

 

Publications

J. Scarlett, I. Bogunovic and V. Cevher. Overlapping Multi-Bandit Best Arm Identification. The 2019 IEEE International Symposium on Information Theory (ISIT), Paris, France, 2019.
I. Bogunovic. Robust Adaptive Decision Making: Bayesian Optimization and Beyond. EPFL, Lausanne, 2019.
I. Bogunovic, J. Scarlett, S. Jegelka and V. Cevher. Adversarially Robust Optimization with Gaussian Processes. Conference on Neural Information Processing Systems (NIPS), Montreal, 2018.
I. Bogunovic, J. Zhao and V. Cevher. Robust Maximization of Non-Submodular Objectives. International Conference on Artificial Intelligence and Statistics (AISTATS), Lanzarote, Canary Islands, 2018.
P. T. Y. Rolland, J. Scarlett, I. Bogunovic and V. Cevher. High Dimensional Bayesian Optimization via Additive Models with Overlapping Groups. AISTATS, Lanzarote, Spain, 2018.
S. Mitrovic, I. Bogunovic, A. Norouzi Fard, J. Tarnawski and V. Cevher. Streaming Robust Submodular Maximization: A Partitioned Thresholding Approach. Conference on Neural Information Processing Systems (NIPS), Long Beach, 2017.
I. Bogunovic, S. Mitrovic, J. Scarlett and V. Cevher. A Distributed Algorithm for Partitioned Robust Submodular Maximization. IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), , 2017.
I. Bogunovic, S. Mitrovic, J. Scarlett and V. Cevher. Robust Submodular Maximization: A Non-Uniform Partitioning Approach. The 34th International Conference on Machine Learning (ICML), Sydney, 2017.
J. Scarlett, I. Bogunovic and V. Cevher. Lower Bounds on Regret for Noisy Gaussian Process Bandit Optimization. Conference on Learning Theory (COLT), Amsterdam, 2017.
V. Cevher, Y.-h. Li, I. Bogunovic, L. Baldassarre and J. Scarlett. Learning-based subsampling, US2017109650, 2017.
A. Norouzi Fard, A. Bazzi, M. El Halabi, I. Bogunovic and Y.-P. Hsieh. An Efficient Streaming Algorithm for the Submodular Cover Problem. The Thirtieth Annual Conference on Neural Information Processing Systems (NIPS), , 2016.
I. Bogunovic, J. Scarlett, A. Krause and V. Cevher. Truncated Variance Reduction: A Unified Approach to Bayesian Optimization and Level-Set Estimation. Conference on Neural Information Processing Systems (NIPS), Barcelona, 2016.
I. Bogunovic, J. Scarlett and V. Cevher. Time-Varying Gaussian Process Bandit Optimization. International Conference on Artificial Intelligence and Statistics (AISTATS), Cadiz, Spain, 2016.
L. Baldassarre, Y.-H. Li, J. Scarlett, B. Gözcü and I. Bogunovic. Learning-Based Compressive Subsampling, in IEEE Journal on Selected Topics in Signal Processing, vol. 10, num. 4, p. 809-822Ieee-Inst Electrical Electronics Engineers Inc, Piscataway, 2016.
I. Bogunovic, V. Cevher, J. Haupt and J. Scarlett. Active Learning of Self-concordant like Multi-index Functions. 40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brisbane, Australia, 2015.
A. Singla, I. Bogunovic, G. Bartok, A. Karbasi and A. Krause. Near-Optimally Teaching the Crowd to Classify . The 31st International Conference on Machine Learning (ICML), Beijing, 2014.
A. Singla, I. Bogunovic, G. Bartok, A. Karbasi and A. Krause. On Actively Teaching the Crowd to Classify. NIPS Workshop on Data Driven Education, , 2013.