A Stochastic Quasi-Newton Method for Online Convex Optimization
نویسندگان
چکیده
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.
منابع مشابه
New Quasi-Newton Optimization Methods for Machine Learning
This thesis develops new quasi-Newton optimization methods that exploit the wellstructured functional form of objective functions often encountered in machine learning, while still maintaining the solid foundation of the standard BFGS quasi-Newton method. In particular, our algorithms are tailored for two categories of machine learning problems: (1) regularized risk minimization problems with c...
متن کاملOn the Behavior of Damped Quasi-Newton Methods for Unconstrained Optimization
We consider a family of damped quasi-Newton methods for solving unconstrained optimization problems. This family resembles that of Broyden with line searches, except that the change in gradients is replaced by a certain hybrid vector before updating the current Hessian approximation. This damped technique modifies the Hessian approximations so that they are maintained sufficiently positive defi...
متن کاملA Variance Reduced Stochastic Newton Method
Quasi-Newton methods are widely used in practise for convex loss minimization problems. These methods exhibit good empirical performance on a wide variety of tasks and enjoy super-linear convergence to the optimal solution. For largescale learning problems, stochastic Quasi-Newton methods have been recently proposed. However, these typically only achieve sub-linear convergence rates and have no...
متن کاملA Self-Correcting Variable-Metric Algorithm for Stochastic Optimization
An algorithm for stochastic (convex or nonconvex) optimization is presented. The algorithm is variable-metric in the sense that, in each iteration, the step is computed through the product of a symmetric positive definite scaling matrix and a stochastic (mini-batch) gradient of the objective function, where the sequence of scaling matrices is updated dynamically by the algorithm. A key feature ...
متن کاملIQN: An Incremental Quasi-Newton Method with Local Superlinear Convergence Rate
The problem of minimizing an objective that can be written as the sum of a set of n smooth and strongly convex functions is challenging because the cost of evaluating the function and its derivatives is proportional to the number of elements in the sum. The Incremental Quasi-Newton (IQN) method proposed here belongs to the family of stochastic and incremental methods that have a cost per iterat...
متن کامل