Kernel methods are among the most popular techniques in machine learning. From a frequentist/discriminative perspective they play a central role in regularization theory as they provide a natural choice for the hypotheses space and the regularization functional through the notion of reproducing kernel Hilbert spaces. From a Bayesian/generative perspective they are the key in the context of Gaussian processes, where the kernel function is also known as the covariance function. Traditionally, kernel methods have been used in supervised learning problem with scalar outputs and indeed there has been a considerable amount of work devoted to designing and learning kernels. More recently there has been an increasing interest in methods that deal with multiple outputs, motivated partly by frameworks like multitask learning. In this paper, we review different methods to design or learn valid kernel functions for multiple outputs, paying particular attention to the connection between probabilistic and functional methods.

We show a principled way of deriving online learning algorithms from a minimax analysis. Various upper bounds on the minimax value, previously thought to be non-constructive, are shown to yield algorithms. This allows us to seamlessly recover known methods and to derive new ones. Our framework also captures such "unorthodox" methods as Follow the Perturbed Leader and the R^2 forecaster. We emphasize that understanding the inherent complexity of the learning problem leads to the development of algorithms. We define local sequential Rademacher complexities and associated algorithms that allow us to obtain faster rates in online learning, similarly to statistical learning theory. Based on these localized complexities we build a general adaptive method that can take advantage of the suboptimality of the observed sequence. We present a number of new algorithms, including a family of randomized methods that use the idea of a "random playout". Several new versions of the Follow-the-Perturbed-Leader algorithms are presented, as well as methods based on the Littlestone's dimension, efficient methods for matrix completion with trace norm, and algorithms for the problems of transductive learning and prediction with static experts.

The interactive partially observable Markov decision process (I-POMDP) is a recently developed framework which extends the POMDP to the multi-agent setting by including agent models in the state space. This paper argues for formulating the problem of an agent learning interactively from a human teacher as an I-POMDP, where the agent \emph{programming} to be learned is captured by random variables in the agent's state space, all \emph{signals} from the human teacher are treated as observed random variables, and the human teacher, modeled as a distinct agent, is explicitly represented in the agent's state space. The main benefits of this approach are: i. a principled action selection mechanism, ii. a principled belief update mechanism, iii. support for the most common teacher \emph{signals}, and iv. the anticipated production of complex beneficial interactions. The proposed formulation, its benefits, and several open questions are presented.

Do quantified and origin) and molecular aspects of our bodies like DNA can tweeting, heavily monitored and selfreporting animals, converge. Consumer genomics websites, crowdsourcing of humans, environments and food create some new biodata but also social networking over genes, together uniformity, a dangerously homogenous, objectified and with services monitoring food flows and food authenticity standardized collective or these data offer some new can create new models of research in nutrigenomics and opportunity for interaction? Are we creating new symbiotic projects related to dieting, health and relations over these data that can lead to a new sense of lifestyle choices. How to connect various scales from community or we are witnessing some depersonalization molecules to institutions and what will be the function of and objectification? How to make meaning out of large these interactions and interfaces? How to create quantities of data and how to bring user experience to data meaningful interaction across scales and large datasets?

The current era of internet-facilitated bigger data, better tools, and collective intelligence community computing is accelerating advances in many areas ranging from artificial intelligence to knowledge generation to public health. In the health sector, data volumes are growing with genomic, phenotypic, microbiomic, metabolomic, self-tracking, and other data streams. Simultaneously, tools are proliferating to allow individuals and groups to make sense of these data in a participatory manner through personal health tracking devices, mobile health applications, and personal electronic medical records. Health community computing models are emerging to support individual activity and mass collaboration through health social networks and crowdsourced health research studies. Participatory health efforts portend important benefits based on both size and speed. Studies can be carried out in cohorts of thousands instead of hundreds, and it could be possible to apply findings from newly-published studies with near-immediate speed. One operator of interventional crowdsourced health research studies, DIYgenomics, has several crowdsourced health research studies in open enrollment as of January 2012 in the areas of vitamin deficiency, aging, mental performance, and epistemology. The farther future of intelligent health community computing could include personal health dashboards, continuous personal health information climates, personal virtual coaches (e.g.; Siri 2.0), and an efficient health frontier of dynamic personalized health recommendations and action-taking.

This paper introduces SIFU, a system that recruits in real time native speakers as online volunteer tutors to help answer questions from Chinese language learners in reading news articles. SIFU integrates the strengths of two effective online language learning methods: reading online news and communicating with online native speakers. SIFU recruits volunteers from an online social network rather than recruits workers from Amazon Mechanical Turk.Initial experiments showed that the proposed approach is able to effectively recruit online volunteer tutors, adequately answer the learners' questions, and efficiently obtain an answer for the learner. Our field deployment illustrates that SIFU is very useful in assisting Chinese learners in reading Chinese news articles and online volunteer tutors are willing to help Chinese learners when they are on social network service.

This paper extends hierarchical task network (HTN) planning with lightweight learning, considering that in robotics, actions have a non-zero probability of failing. Our work applies to A*-based HTN planners with lifting. We prove that the planner finds the plan of maximal expected utility, while retaining its lifting capability and efficient heuristic-based search. We show how to learn the probabilities online, which allows a robot to adapt by replanning on execution failures. The idea behind this work is to use the HTN domain to constrain the space of possibilities, and then to learn on the constrained space in a way requiring few training samples, rendering the method applicable to autonomous mobile robots.

One of the original applications of crowdsourcing the construction of knowledge is Wikipedia, which relies entirely on people to contribute, extend, and modify the representation of knowledge. This paper presents a case for combining AI and wisdom of the crowds for the social construction of knowledge. Our social-computational approach to collective intelligence combines the strengths of human cognitive diversity in producing content and the capabilities of an AI, through methods such as topic modeling, to link and synthesize across these human contributions. In addition to drawing from established domains such as Wikipedia for inspiration and guidance, we present the design of a system that incorporates AI into wisdom of the crowds to develop a knowledge base on sustainability. In this setting the AI plays the role of scholar, as might many of the other participants, drawing connections and synthesizing across contributions. We close with a general discussion, speculating on educational implications and other roles that an AI can play within an otherwise collective human intelligence.

Crowdsourcing requires new strategies to evaluate the workers involved as well as the quality of workers’ output. Using customer feedback data, we introduce multi-facetted Rasch scaling as an evaluation technique to assess the contributions of workers and products simultaneously within a single coherent measurement framework. Based on a data set of about 250,000 customers who rated nearly 115,000 products, for a total of nearly 3 million cases, we found that product ratings reflect almost as much the existence of stable rater differences as they are indicative of the products’ properties. We illustrate how Rasch scaling provides extensive quality control mechanisms; as well we show how aberrant workers and products can be identified so that appropriate feedback and/or corrective actions can be initiated.

Multi-task learning is a learning paradigm which seeks to improve the generalization performance of a learning task with the help of some other related tasks. In this paper, we propose a regularization formulation for learning the relationships between tasks in multi-task learning. This formulation can be viewed as a novel generalization of the regularization framework for single-task learning. Besides modeling positive task correlation, our method, called multi-task relationship learning (MTRL), can also describe negative task correlation and identify outlier tasks based on the same underlying principle. Under this regularization framework, the objective function of MTRL is convex. For efficiency, we use an alternating method to learn the optimal model parameters for each task as well as the relationships between tasks. We study MTRL in the symmetric multi-task learning setting and then generalize it to the asymmetric setting as well. We also study the relationships between MTRL and some existing multi-task learning methods. Experiments conducted on a toy problem as well as several benchmark data sets demonstrate the effectiveness of MTRL.