In most online learning algorithms, the weights assigned to the misclassified examples (or support vectors) remain unchanged during the entire learning process. This is clearly insufficient since when a new misclassified example is added to the pool of support vectors, we generally expect it to affect the weights for the existing support vectors. In this paper, we propose a new online learning method, termed Double Updating Online Learning", or "DUOL" for short. Instead of only assigning a fixed weight to the misclassified example received in current trial, the proposed online learning algorithm also tries to update the weight for one of the existing support vectors. We show that the mistake bound can be significantly improved by the proposed online learning method. Encouraging experimental results show that the proposed technique is in general considerably more effective than the state-of-the-art online learning algorithms."

This paper is based on my invited talk at the 19th Annual Conference on Learning Theory (Pittsburgh, PA, June 24, 2006). In recent years COL T invited talks have tended to aim at establishing connections between the traditio nal concerns of the learning community and the work done by other communities (s uch as game theory, statistics, information theory, and optimization). F ollowing this tradition, I will argue that some ideas from the foundations of prob ability can be fruitfully applied in competitive on-line learning. In this paper I will use the following informal taxonomy of predictions (reminiscent of Shafer's [36], Figure 2, taxonomy of probabilities): D-predictions are mere Decisions. They can never be true or false but can be good or bad.

This means that there is an issue over the best way of evaluating their effectiveness on both sound educational principles and on Human Computer Interface principles. It is the aim of this paper to highlight some of the steps to move toward an objective standard by which to gauge VLEs and ultimately to provide a single overall index measure (essentially a score out of 10) for both usability and educational worth based upon an analysis of accepted standards. An HCI index was constructed for general usability comparison and a separate educational index (EDI index) was designed to provide a measure of educational quality. First the Blackboard VLE and second an open source VLE, Moodle, were tested. As far as possible the open source VLE carried the same content as the Blackboard VLE to allow a comparison of the VLE structure and operation rather than its content. Usability statistics are obtained from a set of standard users.

We establish a relationship between the online mistake-bound model of learning and resource-bounded dimension. This connection is combined with the Winnow algorithm to obtain new results about the density of hard sets under adaptive reductions. This improves previous work of Fu (1995) and Lutz and Zhao (2000), and solves one of Lutz and Mayordomo's "Twelve Problems in Resource-Bounded Measure" (1999).

Issues concerning students involved with online learning paths, that need to be faced by e-Tutors on their day-to-day activity, most often than not fall into known pedagogical patterns - that are problems and difficulties already occurred in the past and dealt with. These pedagogical patterns belong to e-Tutors' know-how and experience and their resolution are frequently a matter of activating routine processes or giving  pre-factored answers; nevertheless statistical data indicates that these issues consume a considerable slice of tutors' time. While a portion of the scientific community is still devoting much effort in developing artificial tutoring systems - by deploying AI/MAS-enabled technologies - the solution being investigated by our team focuses on enhancing already-available, open source LMS by implementing a general-purpose tracking and monitoring toolkit able to support e-Tutors in recognizing and dealing with pedagogical patterns stored into a decentralised Knowledge Base. The system architecture is designed to house multiple platforms (only one adapter interface needs to be written for each LMS) and is able to perform real-time, as well as scheduled, data collection by means of Jade-based agents and schedulers.  Information obtained from the processed data is then returned to the platform via web services and specific interfaces (instant messaging chatbot). The first deployed prototype is currently being experimented in adult higher education learning paths and is able to track student activity, forum readings and writings and offers a basic chat-based help interface. Our aim is to turn a standard LMS into a knowledge aggregator where information about its users, its contents and interactions between the two can be mined via Knowledge Services; resulting data could then be used to refine users' and groups' profiles, to monitor learners' deviance from expected learning path, and ultimately to adjust the applied pedagogical model.

Spacecraft simulation is an integral part of NASA mission planning, real-time mission support, training, and systems engineering. Existing approaches that power these simulations cannot quickly react to the dynamic and complex behavior of the International Space Station (ISS). To address this problem, this paper introduces a unique and efficient method for continuously learning highly accurate models from real-time streaming sensor data, relying on an online learning approach. This approach revolutionizes NASA simulation techniques for space missions by providing models that quickly adapt to real-world feedback without human intervention. A novel regional sliding-window technique for online learning of simulation models is proposed that regionally maintains the most recent data. We also explore a knowledge fusion approach to reduce predictive error spikes when confronted with making predictions in situations that are quite different from training scenarios. We demonstrate substantial error reductions up to 74% in our experimental evaluation on the ISS Electrical Power System and discuss the early deployment of our software in the ISS Mission Control Center (MCC) for ground-based simulations.

We propose a general method called truncated gradient to induce sparsity in the weights of online learning algorithms with convex loss functions. This method has several essential properties: The degree of sparsity is continuous -- a parameter controls the rate of sparsification from no sparsification to total sparsification. The approach is theoretically motivated, and an instance of it can be regarded as an online counterpart of the popular $L_1$-regularization method in the batch setting. We prove that small rates of sparsification result in only small additional regret with respect to typical online learning guarantees. The approach works well empirically. We apply the approach to several datasets and find that for datasets with large numbers of features, substantial sparsity is discoverable.

Our first algorithm, ILK (implicit online learning with kernels), employs a new, implicit update technique that can be applied to a wide variety of convex loss functions. We then introduce a bounded memory version, SILK (sparse ILK), that maintains a compact representation of the predictor without compromising solution quality, even in non-stationary environments. We prove loss bounds and analyze the convergence rate of both. Experimental evidence shows that our proposed algorithms outperform current methods on synthetic and real data.

Our first algorithm, ILK (implicit online learning with kernels), employs a new, implicit update technique that can be applied to a wide variety of convex loss functions. We then introduce a bounded memory version, SILK (sparse ILK), that maintains a compact representation of the predictor without compromising solution quality, even in non-stationary environments. We prove loss bounds and analyze the convergence rate of both. Experimental evidence shows that our proposed algorithms outperform current methods on synthetic and real data.

The learner then updates its parameter vector to minimize a riskfunctional, and the process repeats.