This interdisciplinary journal aims to focus on the exchange of relevant trends and research results as well as the presentation of practical experiences gained while developing and testing elements of technology enhanced learning. So it aims to bridge the gap between pure academic research journals and more practical publications. So it covers the full range from research, application development to experience reports and product descriptions. Readers don't have to pay any fee.
We are living in a new age of widespread remote, online learning. Whether it's homeschool parents turning to online resources to help plan lessons, new families looking for activities for their housebound kids over the summer, or high schoolers looking for additional test prep help, the internet is becoming a virtual classroom for a growing number of kids. And the good news is, the quality of online learning platforms has only grown to meet this demand. Some offer games that teach young children in a fun, engaging way that barely feels like school, while others offer in-depth curriculums in foreign languages for students whose parents only speak one language. So what should you look for when searching for a good online learning platform?
We propose Sketched Online Newton (SON), an online second order learning algorithm that enjoys substantially improved regret guarantees for ill-conditioned data. SON is an enhanced version of the Online Newton Step, which, via sketching techniques enjoys a running time linear in the dimension and sketch size. We further develop sparse forms of the sketching methods (such as Oja's rule), making the computation linear in the sparsity of features. Together, the algorithm eliminates all computational obstacles in previous second order online learning approaches. Papers published at the Neural Information Processing Systems Conference.
By exploiting the duality between boosting and online learning, we present a boosting framework which proves to be extremely powerful thanks to employing the vast knowledge available in the online learning area. Using this framework, we develop various algorithms to address multiple practically and theoretically interesting questions including sparse boosting, smooth-distribution boosting, agnostic learning and, as a by-product, some generalization to double-projection online learning algorithms. Papers published at the Neural Information Processing Systems Conference.