Abstract
We develop stochastic variants of the wellknown BFGS quasi-Newton optimization method, in both full and memory-limited (LBFGS) forms, for online optimization of convex functions. The resulting algorithm performs comparably to a well-tuned natural gradient descent but is scalable to very high-dimensional problems. On standard benchmarks in natural language processing, it asymptotically outperforms previous stochastic gradient methods for parameter estimation in conditional random fields. We are working on analyzing the convergence of online (L)BFGS, and extending it to nonconvex optimization problems.
| Original language | English |
|---|---|
| Pages (from-to) | 436-443 |
| Number of pages | 8 |
| Journal | Journal of Machine Learning Research |
| Volume | 2 |
| Publication status | Published - 2007 |
| Event | 11th International Conference on Artificial Intelligence and Statistics, AISTATS 2007 - San Juan, Puerto Rico Duration: 21 Mar 2007 → 24 Mar 2007 |