Margin Infused Relaxed Algorithm

Margin-infused relaxed algorithm (MIRA) is a machine learning algorithm, an online algorithm for multiclass classification problems. It is designed to learn a set of parameters (vector or matrix) by processing all the given training examples one-by-one and updating the parameters according to each training example, so that the current training example is classified correctly with a margin against incorrect classifications at least as large as their loss. The change of the parameters is kept as small as possible.

A two-class version called binary MIRA simplifies the algorithm by not requiring the solution of a quadratic programming problem (see below). When used in a one-vs.-all configuration, binary MIRA can be extended to a multiclass learner that approximates full MIRA, but may be faster to train.

The flow of the algorithm looks as follows:

Algorithm MIRA
  Input: Training examples $T = \{x_i, y_i\}$
  Output: Set of parameters $w$

  $i$ ← 0, $w^{(0)}$ ← 0
  for $n$ ← 1 to $N$
    for $t$ ← 1 to $|T|$
      $w^{(i+1)}$ ← update $w^{(i)}$ according to $\{x_t, y_t\}$
      $i$ ← $i + 1$
    end for
  end for
  return $\frac{\sum_{j=1}^{N \times |T|} w^{(j)}}{N \times |T|}$

• "←" is a shorthand for "changes to". For instance, "largest ← item" means that the value of largest changes to the value of item.
• "return" terminates the algorithm and outputs the value that follows.

The update step is then formalized as a quadratic programming problem: Find $min\|w^{(i+1)} - w^{(i)}\|$, so that $score(x_t,y_t) - score(x_t,y')\geq L(y_t,y')\ \forall y'$, i.e. the score of the current correct training $y$ must be greater than the score of any other possible $y'$ by at least the loss (number of errors) of that $y'$ in comparison to $y$.

External links

• adMIRAble - MIRA implementation in C++
• Miralium - MIRA implementation in Java
• MIRA implementation for Mahout in Hadoop