• Mathematics, Computer Science
  • Published in ArXiv 2018

Algebraic Machine Learning

@article{MartinMaroto2018AlgebraicML,
  title={Algebraic Machine Learning},
  author={Fernando Martin-Maroto and Gonzalo G. de Polavieja},
  journal={ArXiv},
  year={2018},
  volume={abs/1803.05252}
}
Machine learning algorithms use error function minimization to fit a large set of parameters in a preexisting model. However, error minimization eventually leads to a memorization of the training dataset, losing the ability to generalize to other datasets. To achieve generalization something else is needed, for example a regularization method or stopping the training when error in a validation dataset is minimal. Here we propose a different approach to learning and generalization that is… CONTINUE READING
66
Twitter Mentions

References

Publications referenced by this paper.
SHOWING 1-10 OF 14 REFERENCES

A course in universal algebra

VIEW 6 EXCERPTS
HIGHLY INFLUENTIAL

Complexity of n-Queens Completion

VIEW 1 EXCERPT

Compressed sensing

  • David L. Donoho
  • Mathematics, Computer Science
  • IEEE Transactions on Information Theory
  • 2004
VIEW 1 EXCERPT

Integrated model of visual processing

VIEW 1 EXCERPT