Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Daniel Kroening

Exposing Previously Undetectable Faults in Deep Neural Networks


Jun 01, 2021
Isaac Dunn, Hadrien Pouget, Daniel Kroening, Tom Melham

* Accepted to the ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA 2021) 

  Access Paper or Ask Questions

Compositional Explanations for Image Classifiers


Mar 05, 2021
Hana Chockler, Daniel Kroening, Youcheng Sun


  Access Paper or Ask Questions

Neural Termination Analysis


Feb 07, 2021
Mirco Giacobbe, Daniel Kroening, Julian Parsert


  Access Paper or Ask Questions

Shielding Atari Games with Bounded Prescience


Jan 22, 2021
Mirco Giacobbe, Mohammadhosein Hasanbeig, Daniel Kroening, Hjalmar Wijk

* To appear at AAMAS 2021 

  Access Paper or Ask Questions

Ranking Policy Decisions


Aug 31, 2020
Hadrien Pouget, Hana Chockler, Youcheng Sun, Daniel Kroening


  Access Paper or Ask Questions

Cautious Reinforcement Learning with Logical Constraints


Mar 21, 2020
Mohammadhosein Hasanbeig, Alessandro Abate, Daniel Kroening

* Accepted to AAMAS 2020. arXiv admin note: text overlap with arXiv:1902.00778 

  Access Paper or Ask Questions

Semantic Adversarial Perturbations using Learnt Representations


Jan 29, 2020
Isaac Dunn, Tom Melham, Daniel Kroening


  Access Paper or Ask Questions

CounterExample Guided Neural Synthesis


Jan 25, 2020
Elizabeth Polgreen, Ralph Abboud, Daniel Kroening

* 17 pages, 6 figures 

  Access Paper or Ask Questions

DeepSynth: Program Synthesis for Automatic Task Segmentation in Deep Reinforcement Learning


Nov 22, 2019
Mohammadhosein Hasanbeig, Natasha Yogananda Jeppu, Alessandro Abate, Tom Melham, Daniel Kroening


  Access Paper or Ask Questions

Modular Deep Reinforcement Learning with Temporal Logic Specifications


Sep 23, 2019
Lim Zun Yuan, Mohammadhosein Hasanbeig, Alessandro Abate, Daniel Kroening

* arXiv admin note: text overlap with arXiv:1902.00778 

  Access Paper or Ask Questions

Reinforcement Learning for Temporal Logic Control Synthesis with Probabilistic Satisfaction Guarantees


Sep 11, 2019
Mohammadhosein Hasanbeig, Yiannis Kantaros, Alessandro Abate, Daniel Kroening, George J. Pappas, Insup Lee


  Access Paper or Ask Questions

Explaining Deep Neural Networks Using Spectrum-Based Fault Localization


Aug 06, 2019
Youcheng Sun, Hana Chockler, Xiaowei Huang, Daniel Kroening


  Access Paper or Ask Questions

Generating Realistic Unrestricted Adversarial Inputs using Dual-Objective GAN Training


May 07, 2019
Isaac Dunn, Tom Melham, Daniel Kroening


  Access Paper or Ask Questions

Certified Reinforcement Learning with Logic Guidance


Feb 02, 2019
Mohammadhosein Hasanbeig, Alessandro Abate, Daniel Kroening

* arXiv admin note: text overlap with arXiv:1809.07823 

  Access Paper or Ask Questions

Safety and Trustworthiness of Deep Neural Networks: A Survey


Jan 18, 2019
Xiaowei Huang, Daniel Kroening, Marta Kwiatkowska, Wenjie Ruan, Youcheng Sun, Emese Thamo, Min Wu, Xinping Yi

* The paper requires a significant upgrade, and since it could be completely different with its current version, we believe it is more suitable to withdraw 

  Access Paper or Ask Questions

Logically-Constrained Reinforcement Learning


Oct 22, 2018
Mohammadhosein Hasanbeig, Alessandro Abate, Daniel Kroening


  Access Paper or Ask Questions

Logically-Constrained Neural Fitted Q-Iteration


Sep 20, 2018
Mohammadhosein Hasanbeig, Alessandro Abate, Daniel Kroening


  Access Paper or Ask Questions

Concolic Testing for Deep Neural Networks


Aug 04, 2018
Youcheng Sun, Min Wu, Wenjie Ruan, Xiaowei Huang, Marta Kwiatkowska, Daniel Kroening


  Access Paper or Ask Questions

Global Robustness Evaluation of Deep Neural Networks with Provable Guarantees for L0 Norm


Apr 16, 2018
Wenjie Ruan, Min Wu, Youcheng Sun, Xiaowei Huang, Daniel Kroening, Marta Kwiatkowska

* 39 Pages, Github: https://github.com/L0-TRE/L0-TRE 

  Access Paper or Ask Questions

Testing Deep Neural Networks


Mar 18, 2018
Youcheng Sun, Xiaowei Huang, Daniel Kroening


  Access Paper or Ask Questions