Tomás Pevný

ORCID

Publication Activity (10 Years)

Years Active: 2005-2024
Publications (10 Years): 79

Top Venues

CoRR

IEEE Trans. Inf. Forensics Secur.

WIFS

IH&MMSec

Publications

Milan Papez, Martin Rektoris, Václav Smídl, Tomás Pevný
Sum-Product-Set Networks: Deep Tractable Models for Tree-Structured Graphs. ICLR (2024)
Jiri Nemecek, Tomás Pevný, Jakub Marecek
Generating Likely Counterfactuals Using Sum-Product Networks. CoRR (2024)
Michaela Masková, Matej Zorek, Tomás Pevný, Václav Smídl
Deep anomaly detection on set data: Survey and comparison. Pattern Recognit. 151 (2024)
Florian Merkle, Maximilian Samsinger, Pascal Schöttle, Tomás Pevný
On the Economics of Adversarial Machine Learning. IEEE Trans. Inf. Forensics Secur. 19 (2024)
Jaromír Janisch, Tomás Pevný, Viliam Lisý
Classification with costly features in hierarchical deep sets. Mach. Learn. 113 (7) (2024)
Simon Mandlík, Tomás Pevný, Václav Smídl, Lukás Bajer
Malicious Internet Entity Detection Using Local Graph Inference. IEEE Trans. Inf. Forensics Secur. 19 (2024)
Rony Abecidan, Vincent Itier, Jérémie Boulanger, Patrick Bas, Tomás Pevný
Blind Data Adaptation to tackle Covariate Shift in Operational Steganalysis. CoRR (2024)
Rony Abecidan, Vincent Itier, Jérémie Boulanger, Patrick Bas, Tomás Pevný
Leveraging Data Geometry to Mitigate CSM in Steganalysis. WIFS (2023)
Jiri Nemecek, Tomás Pevný, Jakub Marecek
Improving the Validity of Decision Trees as Explanations. CoRR (2023)
Leah Chrestien, Stefan Edelkamp, Antonín Komenda, Tomás Pevný
Optimize Planning Heuristics to Rank, not to Estimate Cost-to-Goal. NeurIPS (2023)
Leah Chrestien, Tomás Pevný, Stefan Edelkamp, Antonín Komenda
Heuristic Search Optimisation Using Planning and Curriculum Learning Techniques. EPIA (1) (2023)
Quentin Giboulot, Tomás Pevný, Andrew D. Ker
The Non-Zero-Sum Game of Steganography in Heterogeneous Environments. IEEE Trans. Inf. Forensics Secur. 18 (2023)
Jaromír Janisch, Tomás Pevný, Viliam Lisý
NASimEmu: Network Attack Simulator & Emulator for Training Agents Generalizing to Novel Scenarios. CoRR (2023)
Vít Skvára, Tomás Pevný, Václav Smídl
Is AUC the best measure for practical comparison of anomaly detectors? CoRR (2023)
Rony Abecidan, Vincent Itier, Jérémie Boulanger, Patrick Bas, Tomás Pevný
Leveraging Data Geometry to Mitigate CSM in Steganalysis. CoRR (2023)
Leah Chrestien, Tomás Pevný, Stefan Edelkamp, Antonín Komenda
Optimize Planning Heuristics to Rank, not to Estimate Cost-to-Goal. CoRR (2023)
Jaromír Janisch, Tomás Pevný, Viliam Lisý
NASimEmu: Network Attack Simulator & Emulator for Training Agents Generalizing to Novel Scenarios. ESORICS Workshops (2) (2023)
Tomás Pevný, Viliam Lisý, Branislav Bosanský, Petr Somol, Michal Pechoucek
Explaining Classifiers Trained on Raw Hierarchical Multiple-Instance Data. CoRR (2022)
Simon Mandlík, Matej Racinsky, Viliam Lisý, Tomás Pevný
JsonGrinder.jl: automated differentiable neural architecture for embedding arbitrary JSON data. J. Mach. Learn. Res. 23 (2022)
Rony Abecidan, Vincent Itier, Jérémie Boulanger, Patrick Bas, Tomás Pevný
Using Set Covering to Generate Databases for Holistic Steganalysis. WIFS (2022)
Rony Abecidan, Vincent Itier, Jérémie Boulanger, Patrick Bas, Tomás Pevný
Using Set Covering to Generate Databases for Holistic Steganalysis. CoRR (2022)
Solène Bernard, Patrick Bas, John Klein, Tomás Pevný
Backpack: A Backpropagable Adversarial Embedding Scheme. IEEE Trans. Inf. Forensics Secur. 17 (2022)
Lukás Adam, Václav Mácha, Václav Smídl, Tomás Pevný
General framework for binary classification on top samples. Optim. Methods Softw. 37 (5) (2022)
Leah Chrestien, Tomás Pevný, Antonín Komenda, Stefan Edelkamp
A Differentiable Loss Function for Learning Heuristics in A. CoRR (2022)
Vít Skvára, Jan Francå, Matej Zorek, Tomás Pevný, Václav Smídl
Comparison of Anomaly Detectors: Context Matters. IEEE Trans. Neural Networks Learn. Syst. 33 (6) (2022)
Solène Bernard, Patrick Bas, Tomás Pevný, John Klein
Optimizing Additive Approximations of Non-additive Distortion Functions. IH&MMSec (2021)
Maximilian Samsinger, Florian Merkle, Pascal Schöttle, Tomás Pevný
When Should You Defend Your Classifier - A Game-theoretical Analysis of Countermeasures against Adversarial Examples. CoRR (2021)
Simon Mandlík, Tomás Pevný
Mapping the Internet: Modelling Entity Interactions in Complex Heterogeneous Networks. CoRR (2021)
Solène Bernard, Patrick Bas, John Klein, Tomás Pevný
Explicit Optimization of min max Steganographic Game. IEEE Trans. Inf. Forensics Secur. 16 (2021)
Leah Chrestien, Tomás Pevný, Antonín Komenda, Stefan Edelkamp
Heuristic Search Planning with Deep Neural Networks using Imitation, Attention and Curriculum Learning. CoRR (2021)
Simon Mandlík, Matej Racinsky, Viliam Lisý, Tomás Pevný
Mill.jl and JsonGrinder.jl: automated differentiable feature extraction for learning from raw JSON data. CoRR (2021)
Milan Papez, Tomás Pevný, Václav Smídl
Fitting large mixture models using stochastic component selection. CoRR (2021)
Maximilian Samsinger, Florian Merkle, Pascal Schöttle, Tomás Pevný
When Should You Defend Your Classifier? - - A Game-Theoretical Analysis of Countermeasures Against Adversarial Examples. GameSec (2021)
Tomás Pevný, Václav Smídl, Martin Trapp, Ondrej Polácek, Tomás Oberhuber
Sum-Product-Transform Networks: Exploiting Symmetries using Invertible Transformations. PGM (2020)
Tomás Pevný, Václav Smídl, Martin Trapp, Ondrej Polácek, Tomás Oberhuber
Sum-Product-Transform Networks: Exploiting Symmetries using Invertible Transformations. CoRR (2020)
Lukás Adam, Václav Mácha, Václav Smídl, Tomás Pevný
General Framework for Binary Classification on Top Samples. CoRR (2020)
Marek Dedic, Tomás Pevný, Lukás Bajer, Martin Holena
Loss Functions for Clustering in Multi-instance Learning. ITAT (2020)
Vít Skvára, Jan Francu, Matej Zorek, Tomás Pevný, Václav Smídl
Comparison of Anomaly Detectors: Context Matters. CoRR (2020)
Tomás Pevný, Marek Dedic
Nested Multiple Instance Learning in Modelling of HTTP network traffic. CoRR (2020)
Niklas Heim, Tomás Pevný, Václav Smídl
Neural Power Units. NeurIPS (2020)
Jaromír Janisch, Tomás Pevný, Viliam Lisý
Symbolic Relational Deep Reinforcement Learning based on Graph Neural Networks. CoRR (2020)
Jaromír Janisch, Tomás Pevný, Viliam Lisý
Classification with costly features as a sequential decision-making problem. Mach. Learn. 109 (8) (2020)
Niklas Heim, Tomás Pevný, Václav Smídl
Neural Power Units. CoRR (2020)
Martin Kopp, Tomás Pevný, Martin Holena
Anomaly explanation with random forests. Expert Syst. Appl. 149 (2020)
Tomás Pevný, Vojtech Kovarík
Approximation capability of neural networks on spaces of probability measures and tree-structured domains. CoRR (2019)
Paul Prasse, René Knaebel, Lukás Machlica, Tomás Pevný, Tobias Scheffer
Joint detection of malicious domains and infected clients. Mach. Learn. 108 (8-9) (2019)
Niklas Heim, Václav Smídl, Tomás Pevný
Rodent: Relevance determination in ODE. CoRR (2019)
Jaromír Janisch, Tomás Pevný, Viliam Lisý
Deep Reinforcement Learning with Explicitly Represented Knowledge and Variable State and Action Spaces. CoRR (2019)
Jaromír Janisch, Tomás Pevný, Viliam Lisý
Classification with Costly Features Using Deep Reinforcement Learning. AAAI (2019)
Solène Bernard, Tomás Pevný, Patrick Bas, John Klein
Exploiting Adversarial Embeddings for Better Steganography. IH&MMSec (2019)
Paul Prasse, René Knaebel, Lukás Machlica, Tomás Pevný, Tobias Scheffer
Joint Detection of Malicious Domains and Infected Clients. CoRR (2019)
Václav Smídl, Jan Bím, Tomás Pevný
Anomaly scores for generative models. CoRR (2019)
Jaromír Janisch, Tomás Pevný, Viliam Lisý
Classification with Costly Features as a Sequential Decision-Making Problem. CoRR (2019)
Jan Stiborek, Tomás Pevný, Martin Rehák
Probabilistic analysis of dynamic malware traces. Comput. Secur. 74 (2018)
Jan Stiborek, Tomás Pevný, Martin Rehák
Multiple instance learning for malware classification. Expert Syst. Appl. 93 (2018)
Vít Skvára, Tomás Pevný, Václav Smídl
Are generative deep models for novelty detection truly better? CoRR (2018)
Tomás Pevný, Andrew D. Ker
Exploring Non-Additive Distortion in Steganography. IH&MMSec (2018)
Jan Kohout, Tomás Pevný
Network Traffic Fingerprinting Based on Approximated Kernel Two-Sample Test. IEEE Trans. Inf. Forensics Secur. 13 (3) (2018)
Vyacheslav Kungurtsev, Tomás Pevný
Algorithms for solving optimization problems arising from deep neural net models: smooth problems. CoRR (2018)
Vyacheslav Kungurtsev, Tomás Pevný
Algorithms for solving optimization problems arising from deep neural net models: nonsmooth problems. CoRR (2018)
Tomás Pevný, Petr Somol
Using Neural Network Formalism to Solve Multiple-Instance Problems. ISNN (1) (2017)
Jan Stiborek, Tomás Pevný, Martin Rehák
Multiple Instance Learning for Malware Classification. CoRR (2017)
Paul Prasse, Lukás Machlica, Tomás Pevný, Jirí Havelka, Tobias Scheffer
Malware Detection by Analysing Network Traffic with Neural Networks. IEEE Symposium on Security and Privacy Workshops (2017)
Martin Grill, Tomás Pevný, Martin Rehák
Reducing false positives of network anomaly detection by local adaptive multivariate smoothing. J. Comput. Syst. Sci. 83 (1) (2017)
Jaromír Janisch, Tomás Pevný, Viliam Lisý
Classification with Costly Features using Deep Reinforcement Learning. CoRR (2017)
Jakub Lokoc, Tomás Grosup, Premysl Cech, Tomás Pevný, Tomás Skopal
Malware Discovery Using Behaviour-Based Exploration of Network Traffic. SISAP (2017)
Karel Durkota, Viliam Lisý, Christopher Kiekintveld, Karel Horák, Branislav Bosanský, Tomás Pevný
Optimal Strategies for Detecting Data Exfiltration by Internal and External Attackers. GameSec (2017)
Tomás Pevný, Petr Somol
Discriminative models for multi-instance problems with tree-structure. CoRR (2017)
Paul Prasse, Lukás Machlica, Tomás Pevný, Jirí Havelka, Tobias Scheffer
Malware Detection by Analysing Encrypted Network Traffic with Neural Networks. ECML/PKDD (2) (2017)
Tomás Pevný, Martin Kopp, Jakub Kroustek, Andrew D. Ker
Malicons: Detecting Payload in Favicons. Media Watermarking, Security, and Forensics (2016)
Tomás Pevný, Petr Somol
Discriminative Models for Multi-instance Problems with Tree Structure. AISec@CCS (2016)
Andrew D. Ker, Tomás Pevný, Patrick Bas
Rethinking Optimal Embedding. IH&MMSec (2016)
Jan Jusko, Martin Rehák, Jan Stiborek, Jan Kohout, Tomás Pevný
Using Behavioral Similarity for Botnet Command-and-Control Discovery. IEEE Intell. Syst. 31 (5) (2016)
Martin Grill, Tomás Pevný
Learning combination of anomaly detectors for security domain. Comput. Networks 107 (2016)
Tomás Pevný
Loda: Lightweight on-line detector of anomalies. Mach. Learn. 102 (2) (2016)
Tomás Komárek, Martin Grill, Tomás Pevný
Passive NAT detection using HTTP access logs. WIFS (2016)
Premysl Cech, Jan Kohout, Jakub Lokoc, Tomás Komárek, Jakub Marousek, Tomás Pevný
Feature Extraction and Malware Detection on Large HTTPS Data Using MapReduce. SISAP (2016)
Jakub Lokoc, Jan Kohout, Premysl Cech, Tomás Skopal, Tomás Pevný
k-NN Classification of Malware in HTTPS Traffic Using the Metric Space Approach. PAISI (2016)
Tomás Pevný, Petr Somol
Using Neural Network Formalism to Solve Multiple-Instance Problems. CoRR (2016)
Jan Kohout, Tomás Pevný
Unsupervised detection of malware in persistent web traffic. ICASSP (2015)
Tomás Pevný, Ivan Nikolaev
Optimizing pooling function for pooled steganalysis. WIFS (2015)
Jan Kohout, Tomás Pevný
Automatic discovery of web servers hosting similar applications. IM (2015)
Rémi Cogranne, Vahid Sedighi, Jessica J. Fridrich, Tomás Pevný
Is ensemble classifier needed for steganalysis in high-dimensional feature spaces? WIFS (2015)
Tomás Pevný, Andrew D. Ker
Towards dependable steganalysis. Media Watermarking, Security, and Forensics (2015)
Andrew D. Ker, Tomás Pevný
A mishmash of methods for mitigating the model mismatch mess. Media Watermarking, Security, and Forensics (2014)
Andrew D. Ker, Tomás Pevný
The Steganographer is the Outlier: Realistic Large-Scale Steganalysis. IEEE Trans. Inf. Forensics Secur. 9 (9) (2014)
Viliam Lisý, Robert Kessl, Tomás Pevný
Randomized Operating Point Selection in Adversarial Classification. ECML/PKDD (2) (2014)
Tomás Pevný, Andrew D. Ker
Steganographic key leakage through payload metadata. IH&MMSec (2014)
Andrew D. Ker, Patrick Bas, Rainer Böhme, Rémi Cogranne, Scott Craver, Tomás Filler, Jessica J. Fridrich, Tomás Pevný
Moving steganography and steganalysis from the laboratory into the real world. IH&MMSec (2013)
Tomás Pevný, Martin Komon, Martin Rehák
Attacking the IDS learning processes. ICASSP (2013)
Tomás Pevný, Andrew D. Ker
The challenges of rich features in universal steganalysis. Media Watermarking, Security, and Forensics (2013)
Tomás Pevný, Jessica J. Fridrich, Andrew D. Ker
From Blind to Quantitative Steganalysis. IEEE Trans. Inf. Forensics Secur. 7 (2) (2012)
Tomás Pevný
Co-occurrence steganalysis in high dimensions. Media Watermarking, Security, and Forensics (2012)
Tomás Pevný, Martin Rehák, Martin Grill
Detecting anomalous network hosts by means of PCA. WIFS (2012)
Andrew David Ker, Tomás Pevný
Batch steganography in the real world. MM&Sec (2012)
Andrew D. Ker, Tomás Pevný
Identifying a steganographer in realistic and heterogeneous data sets. Media Watermarking, Security, and Forensics (2012)
Patrick Bas, Tomás Filler, Tomás Pevný
"Break Our Steganographic System": The Ins and Outs of Organizing BOSS. Information Hiding (2011)
Tomás Pevný
Detecting messages of unknown length. Media Forensics and Security (2011)
Andrew D. Ker, Tomás Pevný
A new paradigm for steganalysis via clustering. Media Forensics and Security (2011)
Tomás Pevný, Patrick Bas, Jessica J. Fridrich
Steganalysis by subtractive pixel adjacency matrix. IEEE Trans. Inf. Forensics Secur. 5 (2) (2010)

Tomás Pevný.css-eqlu8i{font-size:small;padding-left:4px;}

Publication Activity (10 Years)

Top Topics

Top Venues

Tomás Pevný