Chih-Hong Cheng 鄭志弘


Email: my_family_name(dot)chihhong(at)gmail(dot)com

Short bio:

Chih-Hong Cheng is currently researcher at Fraunhofer IKS, acting as the department head - Safety Assurance for AI. His research interests include software engineering, formal methods, and AI/ML for trustworthy autonomy. He received his doctoral degree in CS from the Technical University of Munich.

Prior employment

  • (2019-2021) Technical manager at DENSO, responsible for the research activities in autonomous driving safety within Europe.

  • (2011-2013, 2015-2019) Permanent researcher at fortiss - Research Institute of the Free State of Bavaria; developed the research topic of dependable AI for autonomous systems.

  • (2013-2015) Scientist in ABB Corporate Research Germany; worked on projects related to intelligent production systems (Industry 4.0), cloud-related technologies for industrial automation (PaaS, IaaS), and the analysis of complex industrial software systems.

Papers (DBLP , Google Scholar):

(copyrights belong to publishers)
  • Mixed-neighborhood, multi-speed cellular automata for safety-aware pedestrian prediction [SEFM'21]

  • Federated learning for driver status monitoring [ITSC'21]

  • Monitoring object detection abnormalities via data-label and post-algorithm abstractions [IROS'21 (pdf)]

  • Safety metrics for semantic segmentation in autonomous driving [AI Test'21 (pdf)]

  • Testing autonomous systems with believed equivalence refinement [AI Test'21 (pdf)]

  • Continuous safety verification of neural networks [DATE'21 (pdf)]

  • Provably robust monitoring of neuron activation patterns [DATE'21 (pdf)]

  • Safety-aware hardening of 3D object detection neural network systems [SAFECOMP'20 (pdf)]

  • Towards robust direct perception networks for automated driving [IV'20 (pdf)]

  • Towards safety verification of direct perception neural networks [DATE'20 (pdf)]

  • nn-dependability-kit: Engineering neural networks for safety-critical autonomous driving systems [ICCAD'19 (pdf)]

  • Runtime monitoring neuron activation patterns [DATE'19 (pdf)]

  • Towards dependability metrics for neural networks [MEMOCODE'18 (pdf)]

  • Quantitative projection coverage for testing ML-enabled autonomous systems [ATVA'18 (pdf)]

  • Verification of binarized neural networks via inter-neuron factoring [VSTTE'18 (pdf)]

  • Neural networks for safety-critical applications - challenges, experiments and perspectives [DATE'18 (pdf)]

  • Maximum resilience of artificial neural networks [ATVA'17 (pdf) (video)]

  • autoCode4: Structural controller synthesis [TACAS'17 (pdf)]

  • Structural synthesis for GXW specifications [CAV'16 (pdf) (video)]

  • Compositional parameter synthesis [FM'16]

  • Semantic degrees for Industrie 4.0 engineering [ESEC/FSE'15 (pdf)]

  • Formal consistency checking over specifications in natural languages [DATE'15 (pdf) (slides)]

  • G4LTL-ST: Automatic generation of PLC programs [CAV'14 (pdf)(video)]

  • JBernstein: A validity checker for generalized polynomial constraints [CAV'13 (pdf)]

  • MGSyn: Automatic synthesis for industrial automation [CAV'12 (pdf) (video1, video2) ]

  • Algorithms for synthesizing priorities in component based systems [ATVA'11 (pdf)]

Talks (slides available for download):

Informal tech writings (in traditional Chinese only):