ALT
2004
Springer
14 years 4 months ago
2004
Springer
Rademacher and Gaussian complexities are successfully used in learning theory for measuring the capacity of the class of functions to be learned. One of the most important propert...
ALT
2005
Springer
14 years 4 months ago
2005
Springer
ALT
2005
Springer
14 years 4 months ago
2005
Springer
ALT
2005
Springer
14 years 4 months ago
2005
Springer
ALT
2005
Springer
14 years 4 months ago
2005
Springer
This paper shows how universal learning can be achieved with expert advice. To this aim, we specify an experts algorithm with the following characteristics: (a) it uses only feedba...
ALT
2005
Springer
2005
Springer
PAC-Learnability of Probabilistic Deterministic Finite State Automata in Terms of Variation Distance
14 years 4 months ago
We consider the problem of PAC-learning distributions over strings, represented by probabilistic deterministic finite automata (PDFAs). PDFAs are a probabilistic model for the gen...
ALT
2005
Springer
14 years 4 months ago
2005
Springer
This paper deals with an unusual phenomenon where most machine learning algorithms yield good performance on the training set but systematically worse than random performance on th...
ALT
2005
Springer
14 years 4 months ago
2005
Springer
In language learning, strong relationships between Gold-style models and query models have recently been observed: in some quite general setting Gold-style learners can be replaced...
ALT
2005
Springer
14 years 4 months ago
2005
Springer
We consider a variant of Gold’s learning paradigm where a learner receives as input n different languages (in form of one text where all input languages are interleaved). Our g...
ALT
2005
Springer
14 years 4 months ago
2005
Springer
Abstract. We describe and analyze an algorithm for predicting a sequence of n-dimensional binary vectors based on a set of experts making vector predictions in [0, 1]n . We measure...