Qiong Gao
Papers
2
Total Citations
54
H-Index
2
About
Qiong Gao is a researcher whose work centers on the foundations of model selection and statistical inference, with a particular emphasis on the Minimum Description Length (MDL) principle. Her major contribution lies in rigorously testing how theoretical MDL—grounded in Kolmogorov complexity—performs in practical model selection, specifically when determining the optimal granularity of a model. Her most-cited paper, "Applying MDL to learn best model granularity" (2000), has garnered 49 citations, establishing a key reference for researchers bridging information theory and machine learning. By demonstrating how MDL can guide the choice of model complexity without overfitting, Gao has helped shape approaches to parsimonious modeling in data science. Her work is notable for translating a provably ideal inference method into actionable practice, offering a principled alternative to heuristic model selection techniques. For students and researchers exploring the intersection of algorithmic information theory and applied statistics, Gao’s contributions provide a clear, empirical validation of MDL’s utility in real-world learning tasks.
Research Focus
Key Achievements
Top Papers
- 1Applying MDL to learn best model granularity49 citations · 2000
- 2Applying MDL to Learning Best Model Granularity5 citations · 2000