We propose a new two stage algorithm LING for large scale regression problems. LING has the same risk as the well known Ridge Regression under the fixed design setting and can be computed much faster. Our experiments have shown that LING performs well in terms of both prediction accuracy and computational efficiency compared with other large scale regression algorithms like Gradient Descent, St...

Natural actor-critics form a popular class of policy search algorithms for finding locally optimal policies for Markov decision processes. In this paper we address a drawback of natural actor-critics that limits their real-world applicability—their lack of safety guarantees. We present a principled algorithm for performing natural gradient descent over a constrained domain. In the context of re...

Recently, important contributions on convergence studies of conjugate gradient methods have been made by Gilbert and Nocedal [6]. They introduce a “sufficient descent condition” to establish global convergence results, whereas this condition is not needed in the convergence analyses of Newton and quasi-Newton methods, [6] hints that the sufficient descent condition, which was enforced by their ...

We propose a class of very simple modifications gradient descent and stochastic leveraging Laplacian smoothing. show that when applied to large variety machine learning problems, ranging from logistic regression deep neural nets, the proposed surrogates can dramatically reduce variance, allow take larger step size, improve generalization accuracy. The methods only involve multiplying usual (sto...

Long Short-Term Memory (LSTM) recurrent neural networks (RNNs) outperform traditional RNNs when dealing with sequences involving not only short-term but also long-term dependencies. The decoupled extended Kalman filter learning algorithm (DEKF) works well in online environments and reduces significantly the number of training steps when compared to the standard gradient-descent algorithms. Prev...

The conjugate gradient optimization algorithm usually used for nonlinear least squares is presented and is combined with the modified back propagation algorithm yielding a new fast training multilayer perceptron (MLP) algorithm (CGFR/AG). The approaches presented in the paper consist of three steps: (1) Modification on standard back propagation algorithm by introducing gain variation term of th...

The long short-term memory (LSTM) network trained by gradient descent solves difficult problems which traditional recurrent neural networks in general cannot. We have recently observed that the decoupled extended Kalman filter training algorithm allows for even better performance, reducing significantly the number of training steps when compared to the original gradient descent training algorit...

Recurrent neural networks (RNN), especially the ones requiring extremely long term memories, are difficult to training. Hence, they provide an ideal testbed for benchmarking the performance of optimization algorithms. This paper reports test results of a recently proposed preconditioned stochastic gradient descent (PSGD) algorithm on RNN training. We find that PSGD may outperform Hessian-free o...