This paper considers the stability of online learning algorithms and its implications for learnability (bounded regret). We introduce a novel quantity called forward regret that intuitively measures how good an online learning algorithm is if it is allowed a one-step look-ahead into the future. We show that given stability, bounded forward regret is equivalent to bounded regret. We also show th...

In recent years there has been growing interest in recognition models using local image features for applications ranging from long range motion matching to object class recognition systems. Currently, many state-of-the-art approaches have models involving very restrictive priors in terms of the number of local features and their spatial relations. The adoption of such priors in those models ar...

Literature suggests that existing learning management systems should be extended to integrate learning activities that aim at enhancing comprehension of students in Computer Science education. Therefore, literature has proposed Computing Augmented Learning Management System (CALMS) that meets the needs of online Computer Science education. However, it’s unclear the current state of architecture...

We propose and develop SG-ELM, a stable online learning algorithm based on stochastic gradients and Extreme Learning Machines (ELM). We propose SG-ELM particularly for systems that are required to be stable during learning; i.e., the estimated model parameters remain bounded during learning. We use a Lyapunov approach to prove both asymptotic stability of estimation error and boundedness in the...

We have recently been studying the case where have a training set T generated from an underlying distribution and our goal is to find some good hypothesis, with respect to the true underlying distribution, using the training set T . We now examine how to use online learning algorithms (which work on individual, arbitrary sequences) in a stochastic setting. Let us consider the training set T as ...

In this paper, a robust online sequential extreme learning machine (ROS-ELM) is proposed. It is based on the original OS-ELM with an adaptive selective ensemble framework. Two novel insights are proposed in this paper. First, a novel selective ensemble algorithm referred to as particle swarm opt imization selective ensemble (PSOSEN) is proposed. Noting that PSOSEN is a general selective ensembl...

Learning algorithms have enjoyed numerous successes in robotic control tasks. In problems with time-varying dynamics, online learning methods have also proved to be a powerful tool for automatically tracking and/or adapting to the changing circumstances. However, for safety-critical applications such as airplane flight, the adoption of these algorithms has been significantly hampered by their l...

Machine learning has been successfully used for target localization in wireless sensor networks (WSNs) due to its accurate and robust estimation against highly nonlinear and noisy sensor measurement. For efficient and adaptive learning, this paper introduces online semi-supervised support vector regression (OSS-SVR). The first advantage of the proposed algorithm is that, based on semi-supervise...

We present a learning algorithm for undiscounted reinforcement learning. Our interest lies in bounds for the algorithm’s online performance after some finite number of steps. In the spirit of similar methods already successfully applied for the exploration-exploitation tradeoff in multi-armed bandit problems, we use upper confidence bounds to show that our UCRL algorithm achieves logarithmic on...

Online learning is discussed from the viewpoint of Bayesian statistical inference. By replacing the true posterior distribution with a simpler parametric distribution, one can define an online algorithm by a repetition of two steps: An update of the approximate posterior, when a new example arrives, and an optimal projection into the parametric family. Choosing this family to be Gaussian, we sh...