Hacker News Reader

Comment by argonaut | Hacker News Reader

Your conclusion is theoretically false.

You can prove that L1 regularization is equivalent to taking the optimal unregularized parameters, setting to parameters below a threshold to 0 (the threshold depends on the regularization parameter), and penalizing the other parameters.

View on HN

Yes, but how do you actually optimize that loss in practice? I'm not saying that a perfect solution with an L1 penalty wouldn't have weights exactly equal to 0. I'm saying that with the optimization techniques that are commonly used, you don't end up with exact zeros.

You're not making sense. If the loss function is convex, adding L1 regularization is still convex. So iterative methods for convex problems (which include SVMs, linear regression, and logistic regression) will find the global optimum.

1) Neural Network Loss functions are not convex. But that isn't the issue here.

2) When you use actual numerical optimization techniques with floating point arithmetic, you don't find an exact minimum (global or local). And you don't get exact zeros.

Have you tried this on a real problem? I wouldn't consider MNIST a real problem, but even there you will not get _exact_ zeros. Try it.

If you rolled your own naive numerical approximation of L1 regularization, you might not have gotten exact zeros. If you use e.g. LARS or cyclical coordinate descent for the L1-regularized parameter cohort, as suited to the problem, you will get exact zeros, as prescribed by the mathematics of L1.

I've never seen anyone optimize a neural network using LARS or cyclical coordinate descent. I thought that's what this entire discussion was about - not arbitrary optimization theory.

Reply on news.ycombinator.com