Suboptimal behavior of Bayes and MDL in classification under misspecification
From MaRDI portal
Publication:2384137
DOI10.1007/s10994-007-0716-7zbMath1470.62084OpenAlexW2078458772WikidataQ117157797 ScholiaQ117157797MaRDI QIDQ2384137
John Langford, Peter D. Grünwald
Publication date: 20 September 2007
Published in: Machine Learning (Search for Journal in Brave)
Full work available at URL: https://doi.org/10.1007/s10994-007-0716-7
Related Items (5)
Generalized Bayes approach to inverse problems with model misspecification ⋮ The no-free-lunch theorems of supervised learning ⋮ Contextuality of misspecification and data-dependent losses ⋮ Between order and chaos: The quest for meaningful information ⋮ NML, Bayes and true distributions: a comment on Karabatsos and Walker (2006)
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- A universal prior for integers and estimation by minimum description length
- On the consistency of Bayes estimates
- Occam's razor
- Inferring decision trees using the minimum description length principle
- Asymptotic behavior of Bayes estimates under possibly incorrect models
- On the stochastic complexity of learning realizable and unrealizable rules
- Some PAC-Bayesian theorems
- Coding decision trees
- Misspecification in infinite-dimensional Bayesian statistics
- 10.1162/153244301753344614
- Merging of Opinions with Increasing Information
- Minimum complexity density estimation
- 10.1162/15324430152748236
- A decision-theoretic extension of stochastic complexity and its applications to learning
- The minimum description length principle in coding and modeling
- 10.1162/153244304773936090
- Minimum Message Length and Kolmogorov Complexity
- Refinements of MDL and MML Coding
- An Information Measure for Classification
This page was built for publication: Suboptimal behavior of Bayes and MDL in classification under misspecification