Multitask learning with expert advice

We consider the problem of prediction with expert advice in the setting where a forecaster is presented with several online prediction tasks. Instead of competing against the best expert separately on each task, we assume the tasks are related, and thus we expect that a few experts will perform well on the entire set of tasks. That is, our forecaster would like, on each task, to compete against the best expert chosen from a small set of experts. While we describe the “ideal” algorithm and its performance bound, we show that the computation required for this algorithm is as hard as computation of a matrix permanent. We present an efficient algorithm based on mixing priors, and prove a bound that is nearly as good for the sequential task presentation case. We also consider a harder case where the task may change arbitrarily from round to round, and we develop an efficient approximate randomized algorithm based on Markov chain Monte Carlo techniques.

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.

Multitask learning with expert advice

We consider the problem of prediction with expert advice in the setting where a forecaster is presented with several online prediction tasks. Instead of competing against the best expert separately on each task, we assume the tasks are related, and thus we expect that a few experts will perform well on the entire set of tasks. That is, our forecaster would like, on each task, to compete against the best expert chosen from a small set of experts. While we describe the "ideal" algorithm and its performance bound, we show that the computation required for this algorithm is as hard as computation of a matrix permanent. We present an efficient algorithm based on mixing priors, and prove a bound that is nearly as good for the sequential task presentation case. We also consider a harder case where the task may change arbitrarily from round to round, and we develop an efficient approximate randomized algorithm based on Markov chain Monte Carlo techniques.

Multitask learning without label correspondences

We propose an algorithm to perform multitask learning where each task has potentially distinct label sets and label correspondences are not readily available. This is in contrast with existing methods which either assume that the label sets shared by different tasks are the same or that there exists a label mapping oracle. Our method directly maximizes the mutual information among the labels, and we show that the resulting objective function can be efficiently optimized using existing algorithms. Our proposed approach has a direct application for data integration with different label spaces, such as integrating Yahoo! and DMOZ web directories.

A dependent partition-valued process for multitask clustering and time evolving network modelling

The fundamental aim of clustering algorithms is to partition data points. We consider tasks where the discovered partition is allowed to vary with some covariate such as space or time. One approach would be to use fragmentation-coagulation processes, but these, being Markov processes, are restricted to linear or tree structured covariate spaces. We define a partition-valued process on an arbitrary covariate space using Gaussian processes. We use the process to construct a multitask clustering model which partitions datapoints in a similar way across multiple data sources, and a time series model of network data which allows cluster assignments to vary over time. We describe sampling algorithms for inference and apply our method to defining cancer subtypes based on different types of cellular characteristics, finding regulatory modules from gene expression data from multiple human populations, and discovering time varying community structure in a social network.

Cross-Domain Multitask Learning with Latent Probit Models

Learning multiple tasks across heterogeneous domains is a challenging problem since the feature space may not be the same for different tasks. We assume the data in multiple tasks are generated from a latent common domain via sparse domain transforms and propose a latent probit model (LPM) to jointly learn the domain transforms, and the shared probit classifier in the common domain. To learn meaningful task relatedness and avoid over-fitting in classification, we introduce sparsity in the domain transforms matrices, as well as in the common classifier. We derive theoretical bounds for the estimation error of the classifier in terms of the sparsity of domain transforms. An expectation-maximization algorithm is derived for learning the LPM. The effectiveness of the approach is demonstrated on several real datasets.

Modelling complex resource requirements in Business Process Management Systems

Real-world business processes are resource-intensive. In work environments human resources usually multitask, both human and non-human resources are typically shared between tasks, and multiple resources are sometimes necessary to undertake a single task. However, current Business Process Management Systems focus on task-resource allocation in terms of individual human resources only and lack support for a full spectrum of resource classes (e.g., human or non-human, application or non-application, individual or teamwork, schedulable or unschedulable) that could contribute to tasks within a business process. In this paper we develop a conceptual data model of resources that takes into account the various resource classes and their interactions. The resulting conceptual resource model is validated using a real-life healthcare scenario.

Multitasking information behaviour in public libraries : a survey study

Multitasking information behaviour is the human ability to handle the demands of multiple information tasks concurrently. When we multitask, we work on two or more tasks and switch between those tasks. Multitasking is the way most of us deal with the complex environment we all live in, and recent studies show that people often engage in multitasking information behaviours. Multitasking information behaviours are little understood, however, and an important area for information behaviour research. Our study investigated the multitasking information behaviours of public library users at the Brentwood and Wilkinsburg Public Libraries in Pittsburgh through diary questionnaires. Findings include that some 63.5 percent of library users engaged in multitasking information behaviours, with a mean of 2.5 topic changes and 2.8 topics per library visit. A major finding of our study is that many people in libraries are seeking information on multiple topics and are engaged in multitasking behaviours. The implications of our findings and further research are also discussed. (Contains 7 tables and 2 figures.)

Pedestrian self-reported exposure to distraction by smart phones while walking and crossing the road

Pedestrian crashes account for approximately 14% of road fatalities in Australia. Crossing the road, while a minor part of total walking, presents the highest crash risk because of potential interaction with motor vehicles. Crash risk is elevated by pedestrian illegal use of the road, which may be widespread (e.g. 20% of crossings at signalised intersections at a sample of sites, Brisbane) and enforcement is rare. Effective road crossing requires integration of multiple skills and judgements, any of which can be hindered by distraction. Observational studies suggest that pedestrians are increasingly likely to ‘multitask’, using mobile technology for entertainment and communication, elevating the risk of distraction while crossing. To investigate this, intercept interviews were conducted with a convenience sample of 211 pedestrians aged 18-65 years in Brisbane CBD. Self-reported frequency of using a smart phone for activities at two levels of distraction: cognitive only (voice calls); or cognitive and visual (text messages, internet access) while walking or crossing the road was collected. Results indicated that smart phone use for potentially distracting activities while walking and while crossing the road was high, especially among 18-30 year olds, who were significantly more likely than 31-44yo or 45-65yo to report smart phone use while crossing the road. For 18-30yo and the higher risk activity of crossing the road, 32% texted at high frequency levels and 27% used internet at high frequency levels. Risky levels of distracted crossing appear to be a growing safety issue for 18-30yo, with greater attention to appropriate interventions needed.

Nonparametric traversability estimation in partially occluded and deformable terrain

Terrain traversability estimation is a fundamental requirement to ensure the safety of autonomous planetary rovers and their ability to conduct long-term missions. This paper addresses two fundamental challenges for terrain traversability estimation techniques. First, representations of terrain data, which are typically built by the rover’s onboard exteroceptive sensors, are often incomplete due to occlusions and sensor limitations. Second, during terrain traversal, the rover-terrain interaction can cause terrain deformation, which may significantly alter the difficulty of traversal. We propose a novel approach built on Gaussian process (GP) regression to learn, and consequently to predict, the rover’s attitude and chassis configuration on unstructured terrain using terrain geometry information only. First, given incomplete terrain data, we make an initial prediction under the assumption that the terrain is rigid, using a learnt kernel function. Then, we refine this initial estimate to account for the effects of potential terrain deformation, using a near-to-far learning approach based on multitask GP regression. We present an extensive experimental validation of the proposed approach on terrain that is mostly rocky and whose geometry changes as a result of loads from rover traversals. This demonstrates the ability of the proposed approach to accurately predict the rover’s attitude and configuration in partially occluded and deformable terrain.

Beyond dataset bias: Multi-task unaligned shared knowledge transfer

Many visual datasets are traditionally used to analyze the performance of different learning techniques. The evaluation is usually done within each dataset, therefore it is questionable if such results are a reliable indicator of true generalization ability. We propose here an algorithm to exploit the existing data resources when learning on a new multiclass problem. Our main idea is to identify an image representation that decomposes orthogonally into two subspaces: a part specific to each dataset, and a part generic to, and therefore shared between, all the considered source sets. This allows us to use the generic representation as un-biased reference knowledge for a novel classification task. By casting the method in the multi-view setting, we also make it possible to use different features for different databases. We call the algorithm MUST, Multitask Unaligned Shared knowledge Transfer. Through extensive experiments on five public datasets, we show that MUST consistently improves the cross-datasets generalization performance. © 2013 Springer-Verlag.

一种FPGA的嵌入式可重构机器人控制系统

提出一种基于FPGA的可重构嵌入式微处理器控制系统.在FPGA中嵌入两个NiosⅡ软核,用VHDL语言编写用户自定义组件.在一个由NiosⅡ软核组成的处理器上实现PWM信号生成、编码器信号处理以及多电机同步伺服运算等,在另一个处理器实现机器人任务管理.该控制系统针对微小型爬壁机器人的控制系统设计,不仅具有良好的实时多任务处理能力,而且具有可重构的特点,因而可应用于一类微小型机器人控制系统以提高其设计的灵活性.

基于有限状态机与Petri网的系统分析与设计

从系统组成、功能需求和体系结构方面介绍了航天器空间对接仿真系统的实时多任务控制系统,基于有限状态机和Petri网方法对其进行了单任务级和多任务级的分析建模,并以此为基础完成系统的详细设计,其中应用分叉和资源共享模型实现了系统的同步和互斥问题。实际应用中应用工程化和模块化的方法完成系统设计,系统运行性能良好。试验证明这种分析设计方法合理可行。

遥控移动式作业机器人实时多任务管理系统

本文介绍遥控移动式作业机器人实时多任务管理系统．该系统是在多总线和位总线相结合的计算机系统上，利用ｉＲＭＸⅡ实时多任务操作系统设计编制的实时多任务管理系统．该系统为遥控机器人提供了数据通讯，人机接口，任务作业等重要管理工作，使机器人能顺利地完成机械手操作和车体运动等重要控制功能。