Optimal strategies and minimax lower bounds for online convex games

A number of learning problems can be cast as an Online Convex Game: on each round, a learner makes a prediction x from a convex set, the environment plays a loss function f, and the learner’s long-term goal is to minimize regret. Algorithms have been proposed by Zinkevich, when f is assumed to be convex, and Hazan et al., when f is assumed to be strongly convex, that have provably low regret. We consider these two settings and analyze such games from a minimax perspective, proving minimax strategies and lower bounds in each case. These results prove that the existing algorithms are essentially optimal.

Matrix regularization techniques for online multitask learning

In this paper we examine the problem of prediction with expert advice in a setup where the learner is presented with a sequence of examples coming from different tasks. In order for the learner to be able to benefit from performing multiple tasks simultaneously, we make assumptions of task relatedness by constraining the comparator to use a lesser number of best experts than the number of tasks. We show how this corresponds naturally to learning under spectral or structural matrix constraints, and propose regularization techniques to enforce the constraints. The regularization techniques proposed here are interesting in their own right and multitask learning is just one application for the ideas. A theoretical analysis of one such regularizer is performed, and a regret bound that shows benefits of this setup is reported.

Closing the gap between bandit and full-information online optimization : high-probability regret bound

We demonstrate a modification of the algorithm of Dani et al for the online linear optimization problem in the bandit setting, which allows us to achieve an O( \sqrt{T ln T} ) regret bound in high probability against an adaptive adversary, as opposed to the in expectation result against an oblivious adversary of Dani et al. We obtain the same dependence on the dimension as that exhibited by Dani et al. The results of this paper rest firmly on those of Dani et al and the remarkable technique of Auer et al for obtaining high-probability bounds via optimistic estimates. This paper answers an open question: it eliminates the gap between the high-probability bounds obtained in the full-information vs bandit settings.

Adaptive online gradient descent

We study the rates of growth of the regret in online convex optimization. First, we show that a simple extension of the algorithm of Hazan et al eliminates the need for a priori knowledge of the lower bound on the second derivatives of the observed functions. We then provide an algorithm, Adaptive Online Gradient Descent, which interpolates between the results of Zinkevich for linear functions and of Hazan et al for strongly convex functions, achieving intermediate rates between [square root T] and [log T]. Furthermore, we show strong optimality of the algorithm. Finally, we provide an extension of our results to general norms.

Online banking customization via tag-based interaction

In this paper, we describe ongoing work on online banking customization with a particular focus on interaction. The scope of the study is confined to the Australian banking context where the lack of customization is evident. This paper puts forward the notion of using tags to facilitate personalized interactions in online banking. We argue that tags can afford simple and intuitive interactions unique to every individual in both online and mobile environments. Firstly, through a review of related literature, we frame our work in the customization domain. Secondly, we define a range of taggable resources in online banking. Thirdly, we describe our preliminary prototype implementation with respect to interaction customization types. Lastly, we conclude with a discussion on future work.

High-Order Concept Associations Mining and Inferential Language Modeling for Online Review Spam Detection

Despite many incidents about fake online consumer reviews have been reported, very few studies have been conducted to date to examine the trustworthiness of online consumer reviews. One of the reasons is the lack of an effective computational method to separate the untruthful reviews (i.e., spam) from the legitimate ones (i.e., ham) given the fact that prominent spam features are often missing in online reviews. The main contribution of our research work is the development of a novel review spam detection method which is underpinned by an unsupervised inferential language modeling framework. Another contribution of this work is the development of a high-order concept association mining method which provides the essential term association knowledge to bootstrap the performance for untruthful review detection. Our experimental results confirm that the proposed inferential language model equipped with high-order concept association knowledge is effective in untruthful review detection when compared with other baseline methods.

High-probability regret bounds for bandit online linear optimization

We present a modification of the algorithm of Dani et al. [8] for the online linear optimization problem in the bandit setting, which with high probability has regret at most O ∗ ( √ T) against an adaptive adversary. This improves on the previous algorithm [8] whose regret is bounded in expectation against an oblivious adversary. We obtain the same dependence on the dimension (n 3/2) as that exhibited by Dani et al. The results of this paper rest firmly on those of [8] and the remarkable technique of Auer et al. [2] for obtaining high probability bounds via optimistic estimates. This paper answers an open question: it eliminates the gap between the high-probability bounds obtained in the full-information vs bandit settings.

Perceptions of online learning in an Australian university : Malaysian students' perspectives

Several researchers have reported that cultural and language differences can affect online interactions and communications between students from different cultural backgrounds. Other researchers have asserted that online learning is a tool that can improve teaching and learning skills, but, its effectiveness depends on how the tool is used. Therefore, this study aims to investigate the kinds of challenges encountered by the international students and how they actually cope with online learning. To date little research exists on the perceptions of online learning environments by international Asian students, in particular Malaysian students who study in Australian Universities; hence this study aims to fill this gap. A mixed-method approach was used to collect quantitative and qualitative data using a modified Online Learning Environment Survey (OLES) instrument and focus group interviews. The sample comprised 76 international students from a university in Brisbane. Thirty-five domestic Australian students were included for comparison. Contrary to assumptions from previous research, the findings revealed that there were few differences between the international Asian students from Malaysia and Australian students with regard to their perceptions of online learning. Another cogent finding that emerged was that online learning was most effective when included within blended learning environments. The students clearly indicated that when learning in a blended environment, it was imperative that appropriate features are blended in and customised to suit the particular needs of international students. The study results indicated that the university could improve the quality of the blended online learning environment by: 1) establishing and maintaining a sense of learning community; 2) enhancing the self motivation of students; and 3) professional development of lecturers/tutors, unit coordinators and learning support personnel. Feedback from focus group interviews, highlighted the students‘ frustration with a lack of cooperative learning, strategies and skills which were expected of them by their lecturers/tutors in order to work productively in groups. They indicated a strong desire for lecturers/tutors to provide them prior training in these strategies and skills. The students identified four ways to optimise learning opportunities in cross-cultural spaces. These were: 1) providing preparatory and ongoing workshops focusing on the dispositions and roles of students within student-centred online learning environments; 2) providing preparatory and ongoing workshops on collaborative group learning strategies and skills; 3) providing workshops familiarising students with Australian culture and language; and 4) providing workshops on strategies for addressing technical problems. Students also indicated a strong desire for professional development of lecturers/tutors focused on: 1) teacher attributes, 2) ways to culturally sensitive curricula, and 3) collaborative learning and cooperative working strategies and skills, and 4) designing flexible program structures. Recommendations from this study will be useful to Australian universities where Asian international students from Malaysia study in blended learning environments. An induction program (online skills, collaborative and teamwork skills, study expectations plus familiarisation with Australian culture) for overseas students at the commencement of their studies; a cultural awareness program for lecturers (cultural sensitivity, ways to communicate and a better understanding of Asian educational systems), upskilling of lecturers‘ ability to structure their teaching online and to apply strong theoretical underpinnings when designing learning activities such as discussion forums, and consistency with regards to how content is located and displayed in a learning management system like Blackboard. Through addressing the research questions in this study, the researcher hopes to contribute to and advance the domain of knowledge related to online learning, and to better understand how international Malaysian students‘ perceive online learning environments. These findings have theoretical and pragmatic significance.