Multiclass learning, boosting, and error-correcting codes
COLT '99 Proceedings of the twelfth annual conference on Computational learning theory
Data mining: practical machine learning tools and techniques with Java implementations
Data mining: practical machine learning tools and techniques with Java implementations
Improved Boosting Algorithms Using Confidence-rated Predictions
Machine Learning - The Eleventh Annual Conference on computational Learning Theory
Information Theory: Coding Theorems for Discrete Memoryless Systems
Information Theory: Coding Theorems for Discrete Memoryless Systems
On the Learnability and Design of Output Codes for Multiclass Problems
COLT '00 Proceedings of the Thirteenth Annual Conference on Computational Learning Theory
Belief Propagation and Revision in Networks with Loops
Belief Propagation and Revision in Networks with Loops
Reducing multiclass to binary: a unifying approach for margin classifiers
The Journal of Machine Learning Research
Unveiling turbo codes: some results on parallel concatenated coding schemes
IEEE Transactions on Information Theory
IEEE Transactions on Information Theory
The capacity of low-density parity-check codes under message-passing decoding
IEEE Transactions on Information Theory
Turbo decoding as an instance of Pearl's “belief propagation” algorithm
IEEE Journal on Selected Areas in Communications
Iterative decoding of compound codes by probability propagation in graphical models
IEEE Journal on Selected Areas in Communications
Introducing the Discriminative Paraconsistent Machine (DPM)
Information Sciences: an International Journal
| Hi-index | 0.00 |
In recent work, we introduced a generalization of ECOC learning under the theory of recursive error correcting codes. We named it RECOC (Recursive ECOC) learning. If long output codewords are allowed, as in the case of problems involving a large number of classes, standard recursive codes such as LDPC or Turbo can be used. However, if the number of classes is moderated, neither good LDPC nor Turbo codes might exist due to the small block lengths involved. In this paper, RECOC learning based on the recently introduced ensemble of Product Accumulated codes is analyzed. Due to their native data storage oriented design, smaller block lengths are allowed. In addition, because of their simplicity, all key concepts regarding RECOC learning can be easily explained.