Learning algorithms


A learning algorithm is a mathematical framework or procedure that calculates the best output given a particular set of data. It does this by updating the calculation based on the difference between the actual and desired output. These algorithms are typically concerned with representation and generalization of the input data.

Latest Research and Reviews

News and Comment

  • News and Views |

    Jeremy Bentham distilled animal behavior as avoiding pain and seeking pleasure. Now, using a combination of different methodological approaches, Roy et al. identify a neural mechanism relevant for learning to avoid pain.

    • Falk Eippert
    •  & Irene Tracey
    Nature Neuroscience 17, 1438–1439
  • News and Views |

    Chaotic networks produce rich temporal dynamics that could be useful for timing, but are extremely sensitive to perturbations. Work now shows that a learning rule for the weights of a chaotic recurrent network can stabilize time-varying activity patterns. This result can be used to train output units to produce generic timed responses.

    • Alfonso Renart
    Nature Neuroscience 16, 772–774