Artificial Intelligence - Special volume on qualitative reasoning about physical systems
Compositional modeling: finding the right model for the job
Artificial Intelligence - Special issue: Qualitative reasoning about physical systems II
ACM Transactions on Mathematical Software (TOMS)
C4.5: programs for machine learning
C4.5: programs for machine learning
Declarative Bias for Specific-to-General ILP Systems
Machine Learning - Special issue on bias evaluation and selection
Machine Learning
Reasoning about nonlinear system identification
Artificial Intelligence
Inducing Process Models from Continuous Data
ICML '02 Proceedings of the Nineteenth International Conference on Machine Learning
Enhancing the Plausibility of Law Equation Discovery
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Inducing hierarchical process models in dynamic domains
AAAI'05 Proceedings of the 20th national conference on Artificial intelligence - Volume 2
Incorporating model identifiability into equation discovery of ODE systems
Proceedings of the 10th annual conference companion on Genetic and evolutionary computation
Constructing explanatory process models from biological data and knowledge
Artificial Intelligence in Medicine
Reducing overfitting in predicting intrinsically unstructured proteins
PAKDD'07 Proceedings of the 11th Pacific-Asia conference on Advances in knowledge discovery and data mining
DS'10 Proceedings of the 13th international conference on Discovery science
Hi-index | 0.00 |
In this paper, we review the paradigm of inductive process modeling, which uses background knowledge about possible component processes to construct quantitative models of dynamical systems. We note that previous methods for this task tend to overfit the training data, which suggests ensemble learning as a likely response. However, such techniques combine models in ways that reduce comprehensibility, making their output much less accessible to domain scientists. As an alternative, we introduce a new approach that induces a set of process models from different samples of the training data and uses them to guide a final search through the space of model structures. Experiments with synthetic and natural data suggest this method reduces error and decreases the chance of including unnecessary processes in the model. We conclude by discussing related work and suggesting directions for additional research.