Architectures that incorporate computing-in-memory (CiM) using emerging nonvolatile memory (NVM) devices have become strong contenders for deep neural network (DNN) acceleration due to their impressive energy efficien...
详细信息
Architectures that incorporate computing-in-memory (CiM) using emerging nonvolatile memory (NVM) devices have become strong contenders for deep neural network (DNN) acceleration due to their impressive energy efficiency. Yet, a significant challenge arises when using these emerging devices: they can show substantial variations during the weight-mapping process. This can severely impact DNN accuracy if not mitigated. A widely accepted remedy for imperfect weight mapping is the iterative write-verify approach, which involves verifying conductance values and adjusting devices if needed. In all existing publications, this procedure is applied to every individual device, resulting in a significant programming time overhead. In our research, we illustrate that only a small fraction of weights need this write-verify treatment for the corresponding devices and the DNN accuracy can be preserved, yielding a notable programming acceleration. Building on this, we introduce U-SWIM, a novel method based on the second derivative. It leverages a single iteration of forward and backpropagation to pinpoint the weights demanding write-verify. Through extensive tests on diverse DNN designs and datasets, U-SWIM manifests up to a 10x programming acceleration against the traditional exhaustive write-verify method, all while maintaining a similar accuracy level. Furthermore, compared to our earlier SWIM technique, U-SWIM excels, showing a 7x speedup when dealing with devices exhibiting nonuniform variations.
Although atomicity plays a key role in data operations of shared variables in parallel computation, researchers haven't treated atomicity in Python in much detail. This study provides a novel approach to integrate...
详细信息
Although atomicity plays a key role in data operations of shared variables in parallel computation, researchers haven't treated atomicity in Python in much detail. This study provides a novel approach to integrate the CPU-based atomic C APIs into Python shared variables by C Foreign Function Interface for Python (CFFI) on all major platforms and utilises Cython to optimise calculation in CPython. Evidence shows that the resulting product, Shared Atomic Enterprise (SAE), could accelerate data operations on shared data types to a large extent. These findings provide a solid evidence base for the massive utilisation of Python atomic operations in parallel computation and concurrent programming.
Gray code, a voltage-level-to-data-bit translation scheme, is widely used in QLC SSDs. However, it causes the four data bits in QLC to exhibit significantly different read and write performance with up to 8 × lat...
详细信息
programming is one of the basic skills necessary for information technology talents. At present, from primary and secondary education to university education, great emphasis is placed on cultivating students' prog...
详细信息
The role of non-formal education in increasing female participation in computerscience (CS) is a hot topic. Short-term interventions, including programming skill outreach activities, have been reported to increase se...
详细信息
ISBN:
(纸本)9798400701399
The role of non-formal education in increasing female participation in computerscience (CS) is a hot topic. Short-term interventions, including programming skill outreach activities, have been reported to increase self efficacy and willingness to pursue computing careers in young women. We explored the impact of a programming outreach activity on three types of measures for 30 female pupils: computing self-efficacy, social participation, and understanding of basic computing concepts. Preliminary results revealed a significant increase in participants' self-efficacy and sense of belonging in computing after the informal learning activity. Students were more focused on tasks when engaging socially with their peers and teachers. A decrease in misconception was observed in uni-structural knowledge but no significant difference was found in multi-structural computing knowledge acquisition. These data provide a baseline for study of the long term impact of outreach activities.
Search/optimization problems are plentiful in scientific and engineering domains. Artificial intelligence has long contributed to the development of search algorithms and declarative programming languages geared towar...
详细信息
Search/optimization problems are plentiful in scientific and engineering domains. Artificial intelligence has long contributed to the development of search algorithms and declarative programming languages geared toward solving and modeling search/optimization problems. Automated reasoning and knowledge representation are the subfields of AI that are particularly vested in these developments. Many popular automated reasoning paradigms provide users with languages supporting optimization statements: answer set programming or MaxSAT or min-one, to name a few. These paradigms vary significantly in their languages and in the ways they express quality conditions on computed solutions. Here we propose a unifying framework of so-called weight systems that eliminates syntactic distinctions between paradigms and allows us to see essential similarities and differences between optimization statements provided by paradigms. This unifying outlook has significant simplifying and explanatory potential in the studies of optimization and modularity in automated reasoning and knowledge representation. It also supplies researchers with a convenient tool for proving the formal properties of distinct frameworks;bridging these frameworks;and facilitating the development of translational solvers.
Due to the importance of spreading computerscience education among young people, we present in this paper our work in preparing and organizing a computerscience competition for children from 8 to 15 years old, named...
详细信息
programming skill is one of the essential basic experience that each student in the field of computerscience has to acquire. To potentially train all students such a skill, teachers should know every student understa...
详细信息
ISBN:
(纸本)9783031428227;9783031428234
programming skill is one of the essential basic experience that each student in the field of computerscience has to acquire. To potentially train all students such a skill, teachers should know every student understanding level during the practice of a programming for individually supporting. Conducting a test is a common method to classify the understanding level of the students. However, it would be a heavy burden for teachers and the student levels are known after the test. The purpose of our study is to classify the understanding level of programming during the practice. In this study, we focus on a block coding learning platform, and we propose a classification method by using mouse tracking heatmaps and machine learning techniques. As a first step of the study, we conduct a test with 18 participants. The results had shown that using mouse click heatmap image and decision tree algorithm was observed to classify students based on their programming logic understanding level through activity on a block coding learning platform. In our future work, we will increase the accuracy of classification and develop a model that can classify the understanding levels almost in real time of during programming practice.
This research paper aims to analyze the strengths and weaknesses associated with the utilization of ChatGPT as an educational tool in the context of undergraduate computerscience education. ChatGPT's usage in tas...
详细信息
ISBN:
(纸本)9798400704239
This research paper aims to analyze the strengths and weaknesses associated with the utilization of ChatGPT as an educational tool in the context of undergraduate computerscience education. ChatGPT's usage in tasks such as solving assignments and exams has the potential to undermine students' learning outcomes and compromise academic integrity. This study adopts a quantitative approach to demonstrate the notable unreliability of ChatGPT in providing accurate answers to a wide range of questions within the field of undergraduate computerscience. While the majority of existing research has concentrated on assessing the performance of Large Language Models in handling programming assignments, our study adopts a more comprehensive approach. Specifically, we evaluate various types of questions such as true/false, multi-choice, multi-select, short answer, long answer, design-based, and coding-related questions. Our evaluation highlights the potential consequences of students excessively relying on ChatGPT for the completion of assignments and exams, including self-sabotage. We conclude with a discussion on how can students and instructors constructively use ChatGPT and related tools to enhance the quality of instruction and the overall student experience.
In the era of the fourth industrial revolution, various internet and communications technologies (ICTs) are being applied to manufacturing systems. Based on these technologies, many companies utilize smart manufacturi...
详细信息
In the era of the fourth industrial revolution, various internet and communications technologies (ICTs) are being applied to manufacturing systems. Based on these technologies, many companies utilize smart manufacturing systems to optimize the design and operation of their lines and to diagnose failures. To build and/or improve production lines, various computer-aided engineering (CAE) tools such as optimization solvers and simulation tools for validation are required. In addition, experts depend on their experience or utilize numerous trial and error processes, implying that a large time investment is required obtain the best layout design, without any guarantee that the result is in fact the best. Therefore, the paper proposes an integrated intelligent layout design framework that automatically derives an optimal layout according the requirements of the layout. The proposed framework uses mixed integer linear programming, simulation-based optimization, and digital twin to perform processes such as assembly line balancing, cell/buffer optimization, and layout planning sequentially and repeatedly to derive an optimal layout. By applying this, it is possible to automatically derive the optimal layout design considering limited resources and physical constraints. In addition, it can contribute to improving productivity and work efficiency at manufacturing sites.
暂无评论