Testing theories with learnable and predictive representations

DOI10.1016/J.JET.2010.07.003zbMATH Open1203.62220OpenAlexW2169220024MaRDI QIDQ617673FDOQ617673

Alvaro Sandroni, Jonathan Weinstein, Nabil I. Al-Najjar, Rann Smorodinsky

Publication date: 13 January 2011

Published in: Journal of Economic Theory (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.jet.2010.07.003

Recommendations

• Expressible inspections
• Nonmanipulable Bayesian testing
• Strategic manipulation of empirical tests
• Comparative Testing of Experts
• Manipulability of Future-Independent Tests

zbMATH Keywords

learningexpert testing

Mathematics Subject Classification ID

Applications of statistics (62P99) Theory of languages and software systems (knowledge-based systems, expert systems, etc.) for artificial intelligence (68T35)

Cites Work

• Testing Multiple Forecasters
• Non-Bayesian Testing of a Stochastic Prediction
• Comparative Testing of Experts
• Manipulability of Future-Independent Tests
• Asymptotic calibration
• Weak and strong merging of opinions
• Approachability in infinite dimensional spaces
• Merging, reputation, and repeated games with incomplete information
• A nonmanipulable test
• The reproducible properties of correct forecasts
• Calibrated forecasting and merging
• Strategic Manipulation of Empirical Tests
• Calibration with Many Checking Rules
• An easier way to calibrate.
• The Complexity of Forecast Testing
• Bayesian Representation of Stochastic Processes under Learning: de Finetti Revisited
• Any Inspection is Manipulable

Cited In (7)

• Item response theory in AI: analysing machine learning classifiers at the instance level
• On comparison of experts
• Mostly calibrated
• Merging and testing opinions
• A prequential test for exchangeable theories
• Nonmanipulable Bayesian testing
• Testing probabilistic equivalence through reinforcement learning

This page was built for publication: Testing theories with learnable and predictive representations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q617673)