Compute the gradient and loss given the features of a single data point, add the gradient to a provided vector to avoid creating new objects, and return loss.
Compute the gradient and loss given the features of a single data point, add the gradient to a provided vector to avoid creating new objects, and return loss.
features for one data point
label for this data point
weights/coefficients corresponding to features
the computed gradient will be added to this vector
loss
Compute the gradient and loss given the features of a single data point.
Compute the gradient and loss given the features of a single data point.
features for one data point
label for this data point
weights/coefficients corresponding to features
(gradient: Vector, loss: Double)
:: DeveloperApi :: Compute gradient and loss for a multinomial logistic loss function, as used in multi-class classification (it is also used in binary logistic regression).
In
The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition
by Trevor Hastie, Robert Tibshirani, and Jerome Friedman, which can be downloaded from http://statweb.stanford.edu/~tibs/ElemStatLearn/ , Eq. (4.17) on page 119 gives the formula of multinomial logistic regression model. A simple calculation shows thatfor K classes multiclass classification problem.
The model weights w = (w_1, w_2, ..., w_{K-1})^T becomes a matrix which has dimension of (K-1) * (N+1) if the intercepts are added. If the intercepts are not added, the dimension will be (K-1) * N.
As a result, the loss of objective function for a single instance of data can be written as
where \alpha(i) = 1 if i != 0, and \alpha(i) = 0 if i == 0, margins_i = x w_i.
For optimization, we have to calculate the first derivative of the loss function, and a simple calculation shows that
where \delta_{i, j} = 1 if i == j, \delta_{i, j} = 0 if i != j, and multiplier = \exp(margins_i) / (1 + \sum_k^{K-1} \exp(margins_i)) - (1-\alpha(y)\delta_{y, i+1})
If any of margins is larger than 709.78, the numerical computation of multiplier and loss function will be suffered from arithmetic overflow. This issue occurs when there are outliers in data which are far away from hyperplane, and this will cause the failing of training once infinity / infinity is introduced. Note that this is only a concern when max(margins) > 0.
Fortunately, when max(margins) = maxMargin > 0, the loss function and the multiplier can be easily rewritten into the following equivalent numerically stable formula.
where sum = \exp(-maxMargin) + \sum_i^{K-1}\exp(margins_i - maxMargin) - 1.
Note that each term, (margins_i - maxMargin) in \exp is smaller than zero; as a result, overflow will not happen with this formula.
For multiplier, similar trick can be applied as the following,
where each term in \exp is also smaller than zero, so overflow is not a concern.
For the detailed mathematical derivation, see the reference at http://www.slideshare.net/dbtsai/2014-0620-mlor-36132297