Data outside the problem domain poses significant threats to the security of AI-based intelligent systems. Aiming to investigate the data domain and out-of-distribution (OOD) data in AI quality management (AIQM) study, this paper proposes to use deep learning techniques for feature representation and develop a novel statistical measure for OOD detection. First, to extract low-dimensional representative features distinguishing normal and OOD data, the proposed research combines the deep auto-encoder (AE) architecture and neuron activation status for feature engineering. Then, using local conditional probability (LCP) in data reconstruction, a novel and superior statistical measure is developed to calculate the score of OOD detection. Experiments and evaluations are conducted on image benchmark datasets and an industrial dataset. Through comparative analysis with other common statistical measures in OOD detection, the proposed research is validated as feasible and effective in OOD and AIQM studies.

Large Language Models (LLMs) have been garnering significant attention of AI researchers, especially following the widespread popularity of ChatGPT. However, due to LLMs' intricate architecture and vast parameters, several concerns and challenges regarding their quality assurance require to be addressed. In this paper, a fine-tuned GPT-based sentiment analysis model is first constructed and studied as the reference in AI quality analysis. Then, the quality analysis related to data adequacy is implemented, including employing the content-based approach to generate reasonable adversarial review comments as the wrongly-annotated data, and developing surprise adequacy (SA)-based techniques to detect these abnormal data. Experi-ments based on Amazon.com review data and a fine-tuned GPT model were implemented. Results were thoroughly discussed from the perspective of AI quality assurance to present the quality analysis of an LLM model on generated adversarial textual data and the effectiveness of using SA on anomaly detection in data quality assurance.

Considering mismatched data have harmful influence on the quality assurance in sentiment analysis, therefore this paper proposed an effective method to detect these mismatches. This study considered data adequacy and model’s confidence together, and proposed to use the surprise adequacy metric for mismatch detection. Experiments were implemented on Amazon.com review data. Performances of mismatched data detection and model retraining were evaluated. The proposed method using Mahalanobis-distance-based surprise adequacy was verified feasible and effective to detect mismatched data in the studied dataset. Moreover, after mismatch detection, retrained model was illustrated useful to improve AI model’s quality.

Deep Learning (DL) based self-driving systems are vigorously developing in big companies. While, as several serious accidents were reported with life- and property-loss, the issue of robustness in DL-based self-driving systems inspires great attention, especially facing with some high-risk cases, like adversarial inputs or corner case scenarios in driving. Considering the existing methods are cumbersome in real driving, therefore this paper proposed a novel and simple way which studies data's uncertainty to rise alarm for manual checking. This method developed a metric describing corner case with respect to DL models, and subsequently evaluated data uncertainty. Experiments on a self-driving system verified the feasibility and usefulness of the proposed method. Code in this paper is released [1].

Technology trends are not only transforming the hardware landscape of end-user devices but are also dramatically changing the types of software applications that are deployed on these devices. With the maturity of cloud computing during the past few years, users increasingly rely on networked applications that are deployed in the cloud. In particular, new applications will emerge where user interactions will be based on real-time continuous media streams instead of the traditional request-response types of interfaces. Furthermore, many of these applications will be multi-user streaming media based interactions instead of a single user interaction with an application. In this paper, we propose a geographic location-aware, hybrid, scalable cloud assisted peer-to-peer (P2P) architecture to support such applications that targets low administration cost, reduced bandwidth consumption, low latency, low initial investment cost and optimized resource usage. The main objective is to develop an efficient media delivery system that leverages locality. We propose a 3-layer novel architecture that uses at the core the cloud for application management, 2-tier edge cloud for supporting geo-dispersed user groups, and at the lowest level peer-to-peer dynamic overlays for locally clustered user groups. The proposed architecture manages multiple streaming sessions simultaneously and each streaming session is an independent entity. Our experiments on PlanetLab show that the dynamic construction and maintenance of delivering streams at both the user-level P2P overlay and edge cloud are indeed feasible and effective.

We succeeded in getting 14.9 TFLOPS performance when running a plasma simulation code IMPACT-3D parallelized with High Performance Fortran on 512 nodes of the Earth Simulator. The theoretical peak performance of the 512 nodes is 32 TFLOPS, which means 45% of the peak performance was obtained with HPF.IMPACT-3D is an implosion analysis code using TVD scheme, which performs three-dimensional compressible and inviscid Eulerian fluid computation with the explicit 5-point stencil scheme for spatial differentiation and the fractional time step for time integration. The mesh size is 2048x2048x4096, and the third dimension was distributed for the parallelization. The HPF system used in the evaluation is HPF/ES, developed for the Earth Simulator by enhancing NEC HPF/SX V2 mainly in communication scalability. Shift communications were manually tuned to get best performance by using HPF/JA extensions, which was designed to give the users more control over sophisticated parallelization and communication optimizations.

Abstract We are developing HPF/SX V2, a High Performance Fortran (HPF) compiler for vector parallel machines. It provides some unique extensions as well as the features of HPF 2.0 and HPF/JA. In particular, this paper describes four of them: (1) the ON directive of HPF 2.0; (2) the REFLECT and LOCAL directives of HPF/JA; (3) vectorization directives; and (4) automatic parallelization. We evaluate these features through some benchmark programs on NEC SX‐5. The results show that each of them achieved a 5–8 times speedup in 8‐CPU parallel execution and the four features are useful for vector parallel execution. We also evaluate the overall performance of HPF/SX V2 by using over 30 well‐known benchmark programs from HPFBench, APR Benchmarks, GENESIS Benchmarks, and NAS Parallel Benchmarks. About half of the programs showed good performance, while the other half suggest weakness of the compiler, especially on its runtimes. It is necessary to improve them to put the compiler to practical use. Copyright © 2002 John Wiley & Sons, Ltd.

Abstract This paper presents a set of extensions on High Performance Fortran (HPF) to make it more usable for parallelizing real‐world production codes. HPF has been effective for programs that a compiler can automatically optimize efficiently. However, once the compiler cannot, there have been no ways for the users to explicitly parallelize or optimize their programs. In order to resolve the situation, we have developed a set of HPF extensions (HPF/JA) to give the users more control over sophisticated parallelization and communication optimizations. They include parallelization of loops with complicated reductions, asynchronous communication, user‐controllable shadow, and communication pattern reuse for irregular remote data accesses. Preliminary experiments have proved that the extensions are effective at increasing HPF's usability. Copyright © 2002 John Wiley & Sons, Ltd.

This paper describes new fast integer sorting methods for single vector and shared-memory parallel vector computers, based on the bucket sort algorithm.Existing vectorization methods for bucket sort have made great efforts to avoid store conflicts of vector scatter operations, and therefore are not so efftcient.The vectorization methods shown in this paper-the retry method, the split vector method and the mask vector method-all actively utilize the nature of the store conflicts to achieve high performance.The parallelization method in this paper uses a feature of shared-memory machines and dynamically changes the partitioning of histogram arrays without any overhead.By combining the retry and the parallelization methods, we got the worlds fastest results for the IS program (Class B) in the NAS Parallel Benchmarks on the NBC $X4.Our methods are also applicable to a wide range of particle simulation programs.