Page 79 - 2017 Brochure
P. 79
研究員
陳郁方 Yu-Fang Chen
Associate Research Fellow
Ph.D., Information Management, National Taiwan University
Tel: +886-2-2788-3799 ext. 1514 Fax: +886-2-2782-4814
Email: yfc@iis.sinica.edu.tw
http://www.iis.sinica.edu.tw/pages/yfc
• Associate Research Fellow, Institute of Information Science, Academia Sinica (2014-present)
• Assistant Research Fellow, Institute of Information Science, Academia Sinica (2009-2014)
• Associate Professor (Adjunct), MIS, National Taipei University (2014-present)
• Assistant Professor (Adjunct), Information Management, National Taiwan University (2010-2014)
• Ph.D., Information Management, National Taiwan University (2009)
• Postdoctoral Researcher, Uppsala University (2009)
• EATCS (European Association for Theoretical Computer Science) award for best theoretical paper at ETAPS 2010.
Research Description For symbolic analysis of string manipulating programs (including
many web-applications), string constraint solvers are the key
My primary research interest is developing theories and algorithms components. We developed an efficient solver of this kind. The core
for ensuring software quality. For example, we study the analysis of our technique is to transform a string constraint to flat automata
of MapReduce programs. In MapReduce, the reducer produces whose language encodes the solutions of the constraints. Our
an output from a list of inputs. Due to the scheduling policy of solver holds the current speed record on the standard Kaluza
the platform, inputs may arrive at the reducers in different orders. benchmarks.
This results in the so called “commutativity problem”, i.e., if the
output of a reducer is independent of the order of its inputs. We Many of the existing approaches for system analysis assume the
proved that the problem is undecidable if the input is a list of existence of a model. However, in reality, models usually need to
mathematical integers, however, the problem becomes decidable be constructed manually, which can be imprecise and prone to
if the input is a list of bounded integers (e.g., the 64-bits integers error. We apply and develop computational learning algorithms to
in computers). We developed an automaton model for reducers, automatically generate models of systems. Compared to manually
named register automata over rational numbers. The model has generated models, models generated from learning algorithms
many good mathematical properties. For example, equivalence and benefit from a statistical guarantee on the similarity to the system to
commutativity problems can be decided in polynomial space. We be analyzed.
plan to use the model as a basis for automatic code generation of
reducers.
Publications 8. Parosh Aziz Abdulla, Mohamed Faouzi Atig, Yu-Fang Chen, Lukás
Holík, Ahmed Rezine, Philipp Rümmer, Jari Stenman “String
1. Yu-Fang Chen, Ondrej Lengal, Tony Tan, Zhilin Wu, “Register Constraints for Verification”, CAV 2014
automata with linear arithmetic”, LICS 2017
9. Yong Li, Yu-Fang Chen, Lijun Zhang and Depeng Liu, “A Novel
2. Yu-Fang Chen, Chih-Duo Hong, Ondrej Lengál, Shin-Cheng Learning Algorithm for Buchi Automata based on Family of DFAs
Mu, Nishant Sinha, Bow-Yaw Wang, “An Executable Sequential and Classification Trees”, TACAS 2017
Specification for Spark Aggregation”, NETYS 2017
10. Yu-Fang Chen, Chiao Hsieh, Ondrej Lengál, Tsung-Ju Lii, Ming-
3. Yu-Fang Chen, Lei Song, Zhilin Wu, “The Commutativity Problem of Hsien Tsai, Bow-Yaw Wang, Farn Wang, “PAC Learning-based
the MapReduce Framework: A Transducer-Based Approach”, CAV Verification and Model Synthesis”, ICSE 2016
2016
77
4. Yu-Fang Chen, Chih-Duo Hong, Nishant Sinha, Bow-Yaw Wang,
“Commutativity of Reducers”, TACAS 2015
5. Parosh Aziz Abdulla, Mohamed Faouzi Atig, Phi-Diep Bui, Yu-Fang
Chen, Lukáš Holík, Ahmed Rezine, Philipp Rümmer, “Flatten and
Conquer (A Framework for Efficient Analysis of String Constraints)”,
PLDI 2017
6. Fang Yu, Ching-Yuan Shueh, Chun-Han Lin, Yu-Fang Chen, Bow-
Yaw Wang, Tevfik Bultan, “Optimal Sanitization Synthesis for Web
Application Vulnerability Repair”, ISSTA 2016
7. Parosh A. Abdulla, Mohamed Faouzi Atig, Yu-Fang Chen, Lukás
Holík, Ahmed Rezine, Philipp Rümmer, Jari Stenman, “Norn: An
SMT Solver for String Constraints”, CAV 2015
陳郁方 Yu-Fang Chen
Associate Research Fellow
Ph.D., Information Management, National Taiwan University
Tel: +886-2-2788-3799 ext. 1514 Fax: +886-2-2782-4814
Email: yfc@iis.sinica.edu.tw
http://www.iis.sinica.edu.tw/pages/yfc
• Associate Research Fellow, Institute of Information Science, Academia Sinica (2014-present)
• Assistant Research Fellow, Institute of Information Science, Academia Sinica (2009-2014)
• Associate Professor (Adjunct), MIS, National Taipei University (2014-present)
• Assistant Professor (Adjunct), Information Management, National Taiwan University (2010-2014)
• Ph.D., Information Management, National Taiwan University (2009)
• Postdoctoral Researcher, Uppsala University (2009)
• EATCS (European Association for Theoretical Computer Science) award for best theoretical paper at ETAPS 2010.
Research Description For symbolic analysis of string manipulating programs (including
many web-applications), string constraint solvers are the key
My primary research interest is developing theories and algorithms components. We developed an efficient solver of this kind. The core
for ensuring software quality. For example, we study the analysis of our technique is to transform a string constraint to flat automata
of MapReduce programs. In MapReduce, the reducer produces whose language encodes the solutions of the constraints. Our
an output from a list of inputs. Due to the scheduling policy of solver holds the current speed record on the standard Kaluza
the platform, inputs may arrive at the reducers in different orders. benchmarks.
This results in the so called “commutativity problem”, i.e., if the
output of a reducer is independent of the order of its inputs. We Many of the existing approaches for system analysis assume the
proved that the problem is undecidable if the input is a list of existence of a model. However, in reality, models usually need to
mathematical integers, however, the problem becomes decidable be constructed manually, which can be imprecise and prone to
if the input is a list of bounded integers (e.g., the 64-bits integers error. We apply and develop computational learning algorithms to
in computers). We developed an automaton model for reducers, automatically generate models of systems. Compared to manually
named register automata over rational numbers. The model has generated models, models generated from learning algorithms
many good mathematical properties. For example, equivalence and benefit from a statistical guarantee on the similarity to the system to
commutativity problems can be decided in polynomial space. We be analyzed.
plan to use the model as a basis for automatic code generation of
reducers.
Publications 8. Parosh Aziz Abdulla, Mohamed Faouzi Atig, Yu-Fang Chen, Lukás
Holík, Ahmed Rezine, Philipp Rümmer, Jari Stenman “String
1. Yu-Fang Chen, Ondrej Lengal, Tony Tan, Zhilin Wu, “Register Constraints for Verification”, CAV 2014
automata with linear arithmetic”, LICS 2017
9. Yong Li, Yu-Fang Chen, Lijun Zhang and Depeng Liu, “A Novel
2. Yu-Fang Chen, Chih-Duo Hong, Ondrej Lengál, Shin-Cheng Learning Algorithm for Buchi Automata based on Family of DFAs
Mu, Nishant Sinha, Bow-Yaw Wang, “An Executable Sequential and Classification Trees”, TACAS 2017
Specification for Spark Aggregation”, NETYS 2017
10. Yu-Fang Chen, Chiao Hsieh, Ondrej Lengál, Tsung-Ju Lii, Ming-
3. Yu-Fang Chen, Lei Song, Zhilin Wu, “The Commutativity Problem of Hsien Tsai, Bow-Yaw Wang, Farn Wang, “PAC Learning-based
the MapReduce Framework: A Transducer-Based Approach”, CAV Verification and Model Synthesis”, ICSE 2016
2016
77
4. Yu-Fang Chen, Chih-Duo Hong, Nishant Sinha, Bow-Yaw Wang,
“Commutativity of Reducers”, TACAS 2015
5. Parosh Aziz Abdulla, Mohamed Faouzi Atig, Phi-Diep Bui, Yu-Fang
Chen, Lukáš Holík, Ahmed Rezine, Philipp Rümmer, “Flatten and
Conquer (A Framework for Efficient Analysis of String Constraints)”,
PLDI 2017
6. Fang Yu, Ching-Yuan Shueh, Chun-Han Lin, Yu-Fang Chen, Bow-
Yaw Wang, Tevfik Bultan, “Optimal Sanitization Synthesis for Web
Application Vulnerability Repair”, ISSTA 2016
7. Parosh A. Abdulla, Mohamed Faouzi Atig, Yu-Fang Chen, Lukás
Holík, Ahmed Rezine, Philipp Rümmer, Jari Stenman, “Norn: An
SMT Solver for String Constraints”, CAV 2015
