Qi Yu

Rochester Institute of Technology | RIT · Department of Information Sciences & Technologies (IST)

Few-shot open-set recognition (FSOSR) aims to detect instances from unseen classes by utilizing a small set of labeled instances from closed-set classes. Accurately rejecting instances from open-set classes in the few-shot setting is fundamentally more challenging due to the weaker supervised signals resulting from fewer labels. Transformer-based f...

Prior neural architecture search (NAS) for adversarial robustness works have discovered that a lightweight and adversarially robust neural network architecture could exist in a non-robust large teacher network, generally disclosed by heuristic rules through statistical analysis and neural architecture search, generally disclosed by heuristic rules...

Ab initio molecular dynamics (AIMD) simulations have become an important tool used in the construction of equations of state (EOS) tables for warm dense matter. Due to computational costs, only a limited number of system state conditions can be simulated, and the remaining EOS surface must be interpolated for use in radiation-hydrodynamic simulatio...

In this paper, we aim to explore novel machine learning (ML) techniques to facilitate and accelerate the construction of universal Equation-Of-State (EOS) models with a high accuracy while ensuring important thermodynamic consistency. When applying ML to fit a universal EOS model, there are two key requirements: (1) a high prediction accuracy to en...

The recently developed sparse network training methods, such as Lottery Ticket Hypothesis (LTH) and its variants, have shown impressive learning capacity by finding sparse sub-networks from a dense one. While these methods could largely sparsify deep networks, they generally focus more on realizing comparable accuracy to dense counterparts yet negl...

Creating accessible software is imperative for making software inclusive for all users. Unfortunately, the topic of accessibility is frequently excluded from computing education, leading to scenarios where students are unaware of either how to develop accessible software or see the need to create it. To address this challenge, we have created a set...

Active learning (AL) aims to sample the most informative data instances for labeling, which makes the model fitting data efficient while significantly reducing the annotation cost. However, most existing AL models make a strong assumption that the annotated data instances are always assigned correct labels, which may not hold true in many practical...

The Conditional Neural Process (CNP) family of models offer a promising direction to tackle few-shot problems by achieving better scalability and competitive predictive performance. However, the current CNP models only capture the overall uncertainty for the prediction made on a target data point. They lack a systematic fine-grained quantification...

Recent years have seen a surge in research on dynamic graph representation learning, which aims to model temporal graphs that are dynamic and evolving constantly over time. However, current work typically models graph dynamics with recurrent neural networks (RNNs), making them suffer seriously from computation and memory overheads on large temporal...

We propose a multimodal data fusion framework to systematically analyze human behavioral data from specialized domains that are inherently dynamic, sparse, and heterogeneous. We develop a two-tier architecture of probabilistic mixtures, where the lower tier leverages parametric distributions from the exponential family to extract significant behavi...

Evidential deep learning, built upon belief theory and subjective logic, offers a principled and computationally efficient way to turn a deterministic neural network uncertainty-aware. The resultant evidential models can quantify fine-grained uncertainty using the learned evidence. To ensure theoretically sound evidential models, the evidence needs...

Open set detection (OSD) aims at identifying data samples of an unknown class (i.e., open set) from those of known classes (i.e., closed set) based on a model trained from closed set samples. However, a closed set may involve a highly imbalanced class distribution. Accurately differentiating open set samples and those from a minority class in the c...

The Conditional Neural Process (CNP) family of models offer a promising direction to tackle few-shot problems by achieving better scalability and competitive predictive performance. However, the current CNP models only capture the overall uncertainty for the prediction made on a target data point. They lack a systematic fine-grained quantification...

Open Set Video Anomaly Detection (OpenVAD) aims to identify abnormal events from video data where both known anomalies and novel ones exist in testing. Unsupervised models learned solely from normal videos are applicable to any testing anomalies but suffer from a high false positive rate. In contrast, weakly supervised methods are effective in dete...

Open Set Video Anomaly Detection (OpenVAD) aims to identify abnormal events from video data where both known anomalies and novel ones exist in testing. Unsupervised models learned solely from normal videos are applicable to any testing anomalies but suffer from a high false positive rate. In contrast, weakly supervised methods are effective in dete...

Multiple-instance learning (MIL) provides an effective way to tackle the video anomaly detection problem by modeling it as a weakly supervised problem as the labels are usually only available at the video level while missing for frames due to expensive labeling cost. We propose to conduct novel Bayesian non-parametric submodular video partition (BN...

As a widely used weakly supervised learning scheme, modern multiple instance learning (MIL) models achieve competitive performance at the bag level. However, instance-level prediction, which is essential for many important applications, remains largely unsatisfactory. We propose to conduct novel active deep multiple instance learning that samples a...

Multiple-instance learning (MIL) provides an effective way to tackle the video anomaly detection problem by modeling it as a weakly supervised problem as the labels are usually only available at the video level while missing for frames due to expensive labeling cost. We propose to conduct novel Bayesian non-parametric submodular video partition (BN...

Optimization-based meta-learning offers a promising direction for few-shot learning that is essential for many real-world computer vision applications. However, learning from few samples introduces uncertainty, and quantifying model confidence for few-shot predictions is essential for many critical domains. Furthermore, few-shot tasks used in meta...

Fine-Grained Sketch-Based Image Retrieval (FG-SBIR) aims at finding a specific image from a large gallery given a query sketch. Despite the widespread applicability of FG-SBIR in many critical domains (e.g., crime activity tracking), existing approaches still suffer from a low accuracy while being sensitive to external noises such as unnecessary st...

Deep learning models have achieved state-of-the-art performance in semantic image segmentation, but the results provided by fully automatic algorithms are not always guaranteed satisfactory to users. Interactive segmentation offers a solution by accepting user annotations on selective areas of the images to refine the segmentation results. However,...

Multiple Instance Learning (MIL) provides a promising solution to many real-world problems, where labels are only available at the bag level but missing for instances due to a high labeling cost. As a powerful Bayesian non-parametric model, Gaussian Processes (GP) have been extended from classical supervised learning to MIL settings, aiming to iden...

Background: One of the most challenging tasks for bladder cancer diagnosis is to histologically differentiate two early stages, non-invasive Ta and superficially invasive T1, the latter of which is associated with a significantly higher risk of disease progression. Indeed, in a considerable number of cases, Ta and T1 tumors look very similar under...

Using web services as building blocks to develop software applications, i.e., service mashups, not only reuses software development efforts to minimize development cost, but also leverages user groups and marketing efforts of those services to attract users and improve profits. This has significantly encouraged the development of a large number of...

In service computing, combining multiple services through service composition to address complex user requirements has become a popular research topic. QoS-aware service composition aims to find the optimal composition scheme with the QoS attributes that best match user requirements. However, certain QoS attributes may continuously change in a dyna...

When self-adaptive systems encounter changes within their surrounding environments, they enact tactics to perform necessary adaptations. For example, a self-adaptive cloud-based system may have a tactic that initiates additional computing resources when response time thresholds are surpassed, or there may be a tactic to activate a specific security...

Studies indicate that much of the software created today is not accessible to all users, indicating that developers don't see the need to devote sufficient resources to creating accessible software. Compounding this problem, there is a lack of robust, easily adoptable educational accessibility material available to instructors for inclusion in thei...

In the service oriented architecture (SOA), software and systems are abstracted as web services to be invoked by other systems. Service composition is a technology, which builds a complex system by combining existing simple services. With the development of SOA and web service technology, massive web services with the same function begin to spring...

Information Technology (IT) and Computer Science (CS) are two well regarded computing programs that produce hundreds of thousands of graduates in each year to meet the diverse needs in the IT industry. Despite being treated as two distinct disciplines, IT and CS do share commonalities as with other computing disciplines (e.g., CE, SE, and IS). In t...

We focus on using natural language unstructured textual Knowledge Bases (KBs) to answer questions from community-based Question-and-Answer (Q8A) websites. We propose a novel framework that integrates multi-level tag recommendation with external KBs to retrieve the most relevant KB articles to answer user posted questions. Different from many existi...

In a service-oriented system, simple services are combined to form value-added services to meet users’ complex requirements. As a result, service composition has become a common practice in service computing. With the rapid development of web service technology, a massive number of web services with the same functionality but different non-function...

System of Systems (SoS) based on service composition is considered as an effective way to build large-scale complex software systems. It regards the system as a service and integrates multiple component systems into a new system. The performance of the component system may fluctuate at any time because of the complex and changeable running state an...

Top-k and skyline techniques have been used to address preference based queries for effective service selection. However, they do not consider the dependencies between attributes in user preferences. In this paper, we focus on developing top-k indexing methods based on Conditional Preference Networks. We first determine whether the correlation amon...

The topological landscape of gene interaction networks provides a rich source of information for inferring functional patterns of genes or proteins. However, it is still a challenging task to aggregate heterogeneous biological information such as gene expression and gene interactions to achieve more accurate inference for prediction and discovery o...

The field of Information Technology (IT) has evolved more rapidly over the past 15 years than ever thought possible. To keep up with industry demands, IT educators have had to react accordingly and with enough foresight to identify those trends that are short lived and those that are here to stay. In this paper, we identify a four-layer IT Stack wi...

With the development of Web service technology, more and more enterprises choose to publish their own services on the Internet. However, with the increasing demands of users, it is difficult for a single service to meet the complex user requirements. To address this challenge, multiple services can be integrated by leveraging the service-oriented a...

Current workflow mining efforts aim to discover process knowledge from user-system interaction logs and represent it as high-level workflow models. They assume there is one single workflow model in a system, or rely on the information that can explicitly link each log sequence to the underlying workflow model. Such assumptions may not be applicable...

The aim of service recommendation is to help a service user find an optimal service. Preference-based service recommendation has gained significant popularity in recent years. Existing studies primarily focus on using similarity measures for quantitative preference. None of them has considered both the quantitative and qualitative preference simult...

Service computing is an emerging technology in System of Systems Engineering (SoS Engineering or SoSE), which regards System as a Service (i.e. SaaS), and aims to construct a robust and value-added complex system by outsourcing external component systems through service composition technology. A service-oriented SoS runs under a dynamic and uncerta...

QoS-aware Web service composition is regarded as one of the fundamental issues in service computing. Given the open and dynamic internet environment, which lacks a central control of individual service providers, we propose in this paper a novel method that seamlessly considers Quality of Service (QoS) and credibility of service providers to achiev...

Service composition provides an effective way to implement a Service-Oriented Architecture (SOA) by combining existing multiple services to meet user requirements. The increasingly complex user requirements and large amount of services pose a significant challenge to service selection and composition. Furthermore, web services are network based, wh...

The rapid proliferation of Internet of Things devices around the world has led to a major increase in demand from industry for students equipped with the skills necessary to make continued advances in this area. Consequently, advanced analytical skills are in urgent need to capitalize the massive amount raw data collected by various IoT devices. To...

With the fast increase of online services of all kinds, users start to care more about the Quality of Service (QoS) that a service provider can offer besides the functionalities of the services. As a result, QoS-based service selection and recommendation have received significant attention since the mid-2000s. However, existing approaches primarily...

Service-oriented architecture is a widely used software engineering paradigm to cope with complexity and dynamics in enterprise applications. Service composition, which provides a cost-effective way to implement software systems, has attracted significant attention from both industry and research communities. As online services may keep evolving ov...

Diagnostic error prevention is a long-established but specialized topic in clinical and psychological research. In this paper, we contribute to the field by exploring diagnostic decision-making via modeling physicians’ utterances of medical concepts during image-based diagnoses. We conduct experiments to collect verbal narratives from derma- tologi...

Mapping out the challenges and strategies for the widespread adoption of service computing.

Image grouping in knowledge-rich domains is challenging, since domain knowledge and human expertise are key to transform image pixels into meaningful content. Manually marking and annotating images is not only labor-intensive but also ineffective. Furthermore, most traditional machine learning approaches cannot bridge this gap for the absence of ex...

Service computing is an emerging technology in System of Systems Engineering (SoS Engineering or SoSE), which regards a System as a Service, and aims at constructing a robust and value-added complex system by outsourcing external component systems through service composition. The burgeoning Big Service computing just covers the significant challeng...

With the increasing popularity of the service-oriented architecture and web service technologies, service composition has become widely adopted to create value-added services from existing ones. As more web services have been deployed on the Internet, it results in a large number of services providing identical functionalities while differing in th...

As a powerful computing paradigm for constructing complex distributed applications, service composition is usually addressed as a planning problem since the goal is to optimize a path for combining services to satisfy special requirements. Some planning methods assume that the state of running environment can be fully observed and monitored. Howeve...

A service-oriented System of Systems (SoS) considers a system as a service and constructs a robust and value-added SoS by outsourcing external component systems through service composition techniques. Online reliability prediction for the component systems for the purpose of assuring the overall Quality of Service (QoS) is often a major challenge i...

This paper reviews and assesses the current coverage of security topics in the master's programs and proposes the best method for educating students in an Information Sciences and Technologies curriculum at Rochester Institute of Technology. We start by describing a case study of student projects in a data-warehousing course to motivate the need to...

Image Grouping in knowledge-rich domains is challenging, since domain knowledge and expertise are key to transform image pix- els into meaningful content. Manually marking and annotating images is not only labor-intensive but also ineffective. On the other hand, pure automated learning technology cannot bridge this gap for the absence of experts’ i...

The articles in this special section aim at presenting the latest developments, trends, and solutions of Big Data analytics on the web.

Service composition offers a powerful software paradigm to build complex and value-added applications by exploiting a service oriented architecture. However, the frequent changes in the internal and external environment demand adaptiveness of a composition solution. Meanwhile, the increasingly complex user requirements and the rapid growth of the c...

Creating mashups from existing Web APIs has provided an effective means to boost software reuse and approach the full potential of online programming resources. One of the key hindrance faced by mashup creation is to discover relevant APIs, especially due to the recent fast growth of Web APIs and the brief, unstructured API descriptions. In this pa...

As online services may keep evolving, service composition should maintain certain adaptivity especially for a dynamic composition environment. Meanwhile, the large number of potential candidate services poses scalability concerns, which demand efficient composition solutions. This paper presents a multi-agent reinforcement learning model for Web se...

The Master of Science (MS) program in Information Sciences and Technologies (IST) at Rochester Institute of Technology conducted a significant upgrade of its curriculum in 2013, aiming to better prepare its graduates for the new trends and challenges in the fast evolving IT computing industry. In particular, the upgraded MS program places a strong...

As the number of Cloud services is growing at a tremendous speed, there is an increasing number of service providers offering similar functionalities. Selecting services with user desired non-functional properties (NFPs) becomes of significant importance but triggers a number of Big Data related research issues. First, the selection decision should...

With the growth of social networks, significant amount of data is brought online that can benefit applications of many kinds if being effectively utilized. As a typical example, Domnigos proposed the concept of viral marketing, which uses the “word of mouth” marketing technique over virtual networks (Domingos, IEEE Intell Syst 20:80–82, 2005). Each...

An organization collects current and historical data for a data warehouse from disparate sources across the organization to support management for making decisions. The data sources change their contents and structure dynamically to reflect business changes or organization requirements, which causes data warehouse evolution in order to provide cons...

Problem solving in complex visual domains involves multiple levels of cognitive processing. Analyzing and representing these cognitive processes requires the elicitation and study of multimodal human data. We have developed methods for extracting experts' visual behaviors and verbal descriptions during medical image inspection. Now we address fusio...

In service computing, online services and the Internet environment are evolving over time, which poses a challenge to service composition for adaptivity. In addition, high efficiency should be maintained when faced with massive candidate services. Consequently, this paper presents a new model for large-scale and adaptive service composition based o...