Key Laboratory of Computer Science,
Institute of Software, Chinese Academy
Email: firstname.lastname@example.org; email@example.com
I obtained a Ph.D. degree in Computer Science from Peking University in July 2012, and graduated with another Ph.D. degree from Griffith University (with National ICT Australia) in July 2014. Then, I joined Institute of Software in Chinese Academy
of Sciences as an Associate Professor, where I was promoted to a Full Professor in 2017.
I have broad interest in artificial intelligence and algorithm engineering, and have particular interest in (1) Practical solving
for NP-hard combinatorial problems, as well as understanding the underlying ideas through both experimental and theoretical analysis; (2) Efficient algorithms for computational logic problems; (3) Automatic algorithm design using optimization and learning techniques.
For published papers, please visit my DBLP
and Google Scholar Citations
Particularly, I would like to introduce Configuration Checking (CC), which is a generic and effective idea for local search. It aims to reduce the cycling phenomenon in local search, by considering the circumstance information (formally defined as configuration) of the variables. It prevents a variable to change its value if its configuration has not changed since the last time it changed value. This idea has witnessed great success in many famous NP hard problems. Here is a report
on CC and its applications.
News on my algorithms/solvers
- In SAT Competition 2018,
ReasonLS placed 1st in
No-Limits track, and placed 2nd in Sparkle SAT Challenge.
- In MaxSAT Evaluation 2018 incomplete solvers track,
SATLike-c placed 1st in
2 categories (out of 4).
- In SAT Competition 2017,
Score2SAT placed 3rd in
- [2016.7.5]: In SAT Competition 2016,
CSCCSAT and DCCAlm placed 2nd and 3rd in
Random track, respectively.
- [2016.7.5]: In MaxSAT Evaluation 2016
incomplete solvers track (9 categories), our solvers placed 1st in 3 categories. The solvers are: CnC-LS, Dist-r, and CCLS. In particular, CnC-LS won the unweighted MaxSAT Industrial category. Note that this category has been dominated by SAT-based solvers for years.
- [2015.9.24]: In MaxSAT Evaluation 2015
incomplete solvers track (9 categories), our solvers placed 1st in 5 categories. The solvers are: DistUP, Dist1, CCLS2015, and CCEHC.
- [2014.7.18]: In SAT Competition 2014,
CCAnr+glucose placed 2nd in
Hard-combinatorial SAT track and 4th in Application SAT track, CSCCSat placed 3rd
in random SAT track.
- [2014.7.18]: In MaxSAT Evaluation 2014
incomplete solvers track (9 categories), Dist placed 1st in 4 categories and CCLS placed
1st in another category.
- [2013.7.12]: In MaxSAT Evaluation
2013, CCLS placed 1st
in 4 categories in "incomplete solvers" track.
- [2012.6.20]: In SAT Challenge 2012
(totally 3 tracks), CCASat placed 1st in Random track.
- [2010.3.20]: EWLS established a new record by finding a
smaller vertex cover (of 3902 vertices) for the frb100-40 challenging problem - details can
be found here.
Some Benchmarks I Maintained
- DIMACS machine benchmark, which has been used to compare speeds of different machines when comparing algorithms for Maximum Clique, Maximum Independent Set, Minimum Vertex Cover and Vertex Coloring.
- Graph Converter, which convert graphs to their complementary graphs.
- DIMACS Clique benchmark of the Second DIMACS Challenge Test Problems (1992-1993).
- DIMACS complementary benchmark, which are complementary graphs of DIMACS Clique benchmark of the Second DIMACS Challenge Test Problems, and are usually used to test algorithms for Maximum Independent Set and Minimum Vertex Cover.
- BHOSLIB benchmark for Graph Problems (Maximum Clique, Maximum Independent Set, Minimum Vertex Cover and Vertex Coloring).
- A collection of undirected real-world large graphs, which contains all the undirected simple graphs in Network Data Repository, and are expressed in DIMACS graph format (a standard graph format in the literature and DIMACS Challenges).
- A collection of weighted real-world large graphs, which were transformed from the above unweighted graphs from Network Data Repository.
PC Member: IJCAI 2016,2017,2018; AAAI 2015,2016,2017,2018
- Junior Associate Editor: Frontiers of Computer Science (2016 Impact Factor: 1.039, SCI indexed), since Dec. 2014
- CCAnr+glucose won the second award in
Hard-combinatorial SAT track, SAT Competition 2014
- CCASat is awarded "Best Sequential Solver" in Random Track,
SAT Challenge 2012
- Distinguished Doctoral Dissertation Award from Peking
- Top-10 Distinguished Academic Fellow Award, EECS of Peking
- Academic Innovation Award, Peking University, 2011
- First Prize in both Medium Sized categories (2vs2 and
4vs4), 2007 RoboCup China Open