A continuous approach to inductive inference
Mathematical Programming: Series A and B
Noise strategies for improving local search
AAAI '94 Proceedings of the twelfth national conference on Artificial intelligence (vol. 1)
An introduction to computational learning theory
An introduction to computational learning theory
Encouraging Experimental Results on Learning CNF
Machine Learning
Artificial Intelligence
The Role of Occam‘s Razor in Knowledge Discovery
Data Mining and Knowledge Discovery
Phase Transitions in Relational Learning
Machine Learning
Analyzing Relational Learning in the Phase Transition Framework
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Further experimental evidence against the utility of Occam's razor
Journal of Artificial Intelligence Research
Domain-independent extensions to GSAT: solving large structured satisfiability problems
IJCAI'93 Proceedings of the 13th international joint conference on Artifical intelligence - Volume 1
An experimental study of phase transitions in matching
IJCAI'99 Proceedings of the 16th international joint conference on Artificial intelligence - Volume 2
Where the really hard problems are
IJCAI'91 Proceedings of the 12th international joint conference on Artificial intelligence - Volume 1
Evidence for invariants in local search
AAAI'97/IAAI'97 Proceedings of the fourteenth national conference on artificial intelligence and ninth conference on Innovative applications of artificial intelligence
C4.5 competence map: a phase transition-inspired approach
ICML '04 Proceedings of the twenty-first international conference on Machine learning
Empirical Study of Relational Learning Algorithms in the Phase Transition Framework
ECML PKDD '09 Proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases: Part I
| Hi-index | 0.00 |
In the past decade, there has been a lot of interest in phase transitions within artificial intelligence, and more recently, in machine learning and inductive logic programming. We investigate phase transitions in learning k-term DNF boolean formulae, a practically relevant class of concepts. We do not only show that there exist phase transitions, but also characterize and locate these phase transitions using the parameters k, the number of positive and negative examples, and the number of boolean variables. Subsequently, we investigate stochastic local search (SLS) for k-term DNF learning. We compare several variants that first reduce k-term DNF to SAT and then apply well-known SLS algorithms, such as GSAT and WalkSAT. Our experiments indicate that WalkSAT is able to solve the largest fraction of hard problem instances.