The aim of multi-output learning is to simultaneously predict multiple outputs given an input. It important problem for decision-making since making decisions in the real world often involves complex factors and criteria. In recent times, increasing number research studies have focused on ways at once. Such efforts transpired different forms according particular under study. Classic cases include multi-label learning, multi-dimensional multi-target regression, others. From our survey topic,...

Multi-output learning with the task of simultaneously predicting multiple outputs for an input has increasingly attracted interest from researchers due to its wide application. The <inline-formula><tex-math notation="LaTeX">$k$</tex-math></inline-formula> nearest neighbor ( notation="LaTeX">$k \text{NN}$</tex-math></inline-formula> ) algorithm is one most popular frameworks handling multi-output problems. performance depends crucially on metric used compute distance between different...

Multi-output learning aims to simultaneously predict multiple outputs given an input. It is important problem due the pressing need for sophisticated decision making in real-world applications. Inspired by big data, 4Vs characteristics of multi-output imposes a set challenges learning, terms volume, velocity, variety and veracity outputs. Increasing number works literature have been devoted study development novel approaches addressing encountered. However, it lacks comprehensive overview on...

Approximate nearest neighbor (ANN) search has achieved great success in many tasks. However, existing popular methods for ANN search, such as hashing and quantization methods, are designed static databases only. They cannot handle well the database with data distribution evolving dynamically, due to high computational effort retraining model based on new database. In this paper, we address problem by developing an online product (online PQ) incrementally updating codebook that accommodates...

Continuous delivery and deployment are dramatically shortening release cycles from months to hours. Cloud applications with high-frequency releases often rely heavily on automated tools cloud infrastructure APIs deploy new software versions. The authors report reliability issues how these contribute them. They also analyze the trade-offs between using baked lightly virtual-image approaches, basis of experiments Amazon Web Service OpsWorks Chef configuration management tool. Finally, they...

Enterprise resilience plays an important role to prevent business services from disruptions caused by human-induced disasters such as failed change implementations and software bugs. Traditional expert-centric approach has difficulty maintain continued critical functions because the can often only be handled after their occurrence. This paper introduces a data-analytics approach, which leverages system monitoring data for enterprise resilience. With power of mining machine learning...

Conducting (big) data analytics in an organization is not just about using a processing framework (e.g. Hadoop/Spark) to learn model from currently single file system HDFS). We frequently need pipeline real time other systems into the framework, and continually update learned model. The frameworks be easily invokable for different purposes produce models. subsequent updates integrated with product that may require prediction latest trained All these shared among teams purposes. In this...

Although content sharing provides many benefits, owners lose full control of their once they are given away. Existing solutions provide limited capabilities access as vendor-specific, non-structured and non-flexible. In this paper, we present an open flexible software solution called SelfProtect Object (SPO). SPO bundles policy files in object that can protect its contents by itself anywhere anytime. Our is based on XACML, a generic language allowing fine-grain with rules conditions. We also...

Label embedding plays an important role in many real-world applications. To enhance the label relatedness captured by embeddings, multiple contexts can be adopted. However, these are heterogeneous and often partially observed practical tasks, imposing significant challenges to capture overall among labels. In this paper, we propose a general Partial Heterogeneous Context Embedding (PHCLE) framework address challenges. Categorizing into two groups, relational context descriptive context,...

The anatomical structure of the brain can be observed via non-invasive techniques such as diffusion imaging. However, these are imperfect because they miss connections that actually known to exist, especially long range inter-hemispheric ones. In this paper we formulate inverse problem inferring structural connectivity networks from experimentally functional Magnetic Resonance Imaging (fMRI), by formulating it a convex optimization problem. We show modeled an optimal sparse representation...

Multiview alignment, achieving one-to-one correspondence of multiview inputs, is critical in many real-world applications, especially for cross-view data analysis problems. An increasing amount work has studied this alignment problem with canonical correlation (CCA). However, existing CCA models are prone to misalign the multiple views due either neglect uncertainty or inconsistent encoding views. To tackle these two issues, letter studies from a Bayesian perspective. Delving into...

Label embedding plays an important role in many real-world applications. To enhance the label relatedness captured by embeddings, multiple contexts can be adopted. However, these are heterogeneous and often partially observed practical tasks, imposing significant challenges to capture overall among labels. In this paper, we propose a general Partial Heterogeneous Context Embedding (PHCLE) framework address challenges. Categorizing into two groups, relational context descriptive context,...

Canonical Correlation Analysis (CCA) is a classic technique for multi-view data analysis. To overcome the deficiency of linear correlation in practical learning tasks, various CCA variants were proposed to capture nonlinear dependency. However, it non-trivial have an in-principle understanding these due their inherent restrictive assumption on and latent code distributions. Although some works studied probabilistic interpretation CCA, models still require explicit form distributions achieve...

Multi-view alignment, achieving one-to-one correspondence of multi-view inputs, is critical in many real-world applications, especially for cross-view data analysis problems. Recently, an increasing number works study this alignment problem with Canonical Correlation Analysis (CCA). However, existing CCA models are prone to misalign the multiple views due either neglect uncertainty or inconsistent encoding views. To tackle these two issues, paper studies from Bayesian perspective. Delving...