Aggregating Algorithm for prediction of packs
| Autor: | Dmitry Adamskiy, Yuri Kalnishkan, Tony Bellotti, Raisa Dzhamtyrova |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
FOS: Computer and information sciences
Technology Computer science 1702 Cognitive Sciences Expert advice House prices 02 engineering and technology Computer Science Artificial Intelligence Machine Learning (cs.LG) Prediction with expert advice 68Q32 68T05 Artificial Intelligence 020204 information systems Machine learning 0801 Artificial Intelligence and Image Processing 0202 electrical engineering electronic engineering information engineering Artificial Intelligence & Image Processing Special case Protocol (object-oriented programming) Sport Science & Technology On-line learning Computer Science - Learning House price Order (business) Computer Science 020201 artificial intelligence & image processing Algorithm Software |
| Zdroj: | Machine Learning. 108:1231-1260 |
| ISSN: | 1573-0565 0885-6125 |
| Popis: | This paper formulates a protocol for prediction of packs, which is a special case of on-line prediction under delayed feedback. Under the prediction of packs protocol, the learner must make a few predictions without seeing the respective outcomes and then the outcomes are revealed in one go. The paper develops the theory of prediction with expert advice for packs by generalising the concept of mixability. We propose a number of merging algorithms for prediction of packs with tight worst case loss upper bounds similar to those for Vovk’s Aggregating Algorithm. Unlike existing algorithms for delayed feedback settings, our algorithms do not depend on the order of outcomes in a pack. Empirical experiments on sports and house price datasets are carried out to study the performance of the new algorithms and compare them against an existing method. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink