CV

Education

• (2017) PhD in discrete optimization from IST Austria, under supervision of Vladimir Kolmogorov
• (2012) Master’s degree in Functional analysis from Charles University, Prague, under supervision of Jiří Spurný

Work experience

• (2017-) Postdoctoral researcher in Autonomous Learning group at Max-Planck-Institute for Intelligent Systems, in Tubingen, Germany.
• (2017) Research engineer for cryptographic startup Wickr
• (summer 2016) PhD Internhsip at Google inc. Data analysis of DoubleClick customer revenue.

Skills

• Python, C++, Linux, Git, scientific computing, PyTorch, TensorFlow, LaTeX

Awards

• CVPR 2020 best paper award nomination
• CVPR 2018 HM best paper award
• Czech Republic Mathematical Olympiad absolute winner (2007)
• International Mathematical Olympiad bronze medal (2007)

Service and leadership

• Reviewer for NeurIPS, CVPR.
• Organizer of mathematical competitions
• Occasional teaching in Africa

Publications

Sample-efficient Cross-Entropy Method for Real-time Planning

Deep Graph Matching via Blackbox Differentiation of Combinatorial Solvers

Optimizing Rank-based Metrics with Blackbox Differentiation

Differentiation of Blackbox Combinatorial Solvers

Variational Autoencoders Pursue PCA Directions (by Accident)

L4: Practical loss-based stepsize adaptation for deep learning

Efficient Optimization for Rank-based Loss Functions

Even Delta-Matroids and the Complexity of Planar Boolean CSPs

On the Memory-Hardness of Data-Independent Password-Hashing Functions

Total variation on a tree

The Complexity of General-Valued CSPs

Effectiveness of Structural Restrictions for Hybrid CSPs

Superconcentrators of Density 25.3