A Deterministic Analysis of an Online Convex Mixture of Experts Algorithm

Abstract

We analyze an online learning algorithm that adaptively combines outputs of two constituent algorithms (or the experts) running in parallel to estimate an unknown desired signal. This online learning algorithm is shown to achieve and in some cases outperform the mean-square error (MSE) performance of the best constituent algorithm in the steady state. However, the MSE analysis of this algorithm in the literature uses approximations and relies on statistical models on the underlying signals. Hence, such an analysis may not be useful or valid for signals generated by various real-life systems that show high degrees of nonstationarity, limit cycles and that are even chaotic in many cases. In this brief, we produce results in an individual sequence manner. In particular, we relate the time-accumulated squared estimation error of this online algorithm at any time over any interval to the one of the optimal convex mixture of the constituent algorithms directly tuned to the underlying signal in a deterministic sense without any statistical assumptions. In this sense, our analysis provides the transient, steady-state, and tracking behavior of this algorithm in a strong sense without any approximations in the derivations or statistical assumptions on the underlying signals such that our results are guaranteed to hold. We illustrate the introduced results through examples.

Low Regret Binary Sampling Method for Efficient Global Optimization of Univariate Functions

“…Over the past years, the global optimization problem has gathered significant attention with various algorithms being proposed in distinct fields of research. It has been studied especially in the fields of non-convex optimization [6]- [8], Bayesian optimization [9], convex optimization [10]- [12], bandit optimization [13], stochastic optimization [14], [15]; because of its practical applications in distribution estimation [16]- [19], multi-armed bandits [20]- [22], control theory [23], signal processing [24], game theory [25], prediction [26], [27], decision theory [28] and anomaly detection [29]- [31].…”

Efficient, Anytime Algorithms for Calibration with Isotonic Regression under Strictly Convex Losses

“…In the problems of learning, recognition, estimation or prediction [1]- [3]; decisions are often produced to minimize certain loss functions using features of the observations, which are generally noisy, random or even missing. There are numerous applications in a number of varying fields such as decision theory [4], control theory [5], game theory [6], [7], optimization [8], [9], density estimation and anomaly detection [10]- [15], scheduling [16], signal processing [17], [18], forecasting [19], [20] and bandits [21]- [23]. These decisions are acquired from specific learning models, where the goal is to distinguish certain data patterns and provide accurate estimations for practical use.…”

The Regression Learning of the Imbalanced and Big Data by the Online Mixture Model for the Mach Number Forecasting

“…Then, the subtask models are learnt in parallel on the data from different working conditions by the same or diverse learning algorithms. The prediction of a query sample (only with input variables) by an OMM is the output from the subtask model that the query sample belongs to [29]- [31].…”