检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Singh, Siddharth Singhania, Prajwal Ranjan, Aditya Kirchenbauer, John Geiping, Jonas Wen, Yuxin Jain, Neel Hans, Abhimanyu Shu, Manli Tomar, Aditya Goldstein, Tom Bhatele, Abhinav Department of Computer Science University of Maryland College Park United States ELLIS Institute Tübingen Max Planck Institute for Intelligent Systems Tübingen Germany Department of Electrical Engineering and Computer Sciences University of California Berkeley United States

Training and fine-tuning large language models (LLMs) with hundreds of billions to trillions of parameters requires tens of thousands of GPUs, and a highly scalable software stack. In this work, we present a novel four-dimensional hybrid parallel algorithm implemented in a highly scalable, portable, open-source framework called AxoNN. We describe several performance optimizations in AxoNN to improve matrix multiply kernel performance, overlap non-blocking collectives with computation, and performance modeling to choose performance optimal configurations. These have resulted in unprecedented scaling and peak flop/s (bf16) for training of GPT-style transformer models on Perlmutter (620.1 Petaflop/s), Frontier (1.381 Exaflop/s) and Alps (1.423 Exaflop/s). While the abilities of LLMs improve with the number of trainable parameters, so do privacy and copyright risks caused by memorization of training data, which can cause disclosure of sensitive or private information at inference time. We highlight this side effect of scale through experiments that explore "catastrophic memorization," where models are sufficiently large to memorize training data in a single pass, and present an approach to prevent it. As part of this study, we demonstrate fine-tuning of a 405-billion parameter LLM using AxoNN on Frontier. Copyright © 2025, The Authors. All rights reserved.

关键词： Supercomputers

来源：评论

学校读者我要写书评

暂无评论

ON THE MAXIMUM SIZE OF VARIABLE-LENGTH NON-OVERLAPPING CODES

arXiv

引用

arXiv 2024年

作者： Wang, Geyang Wang, Qi Department of Electrical and Computer Engineering Institute for Systems Research University of Maryland College ParkMD20742 United States Department of Computer Science and Engineering National Center for Applied Mathematics Shenzhen Southern University of Science and Technology Nanshan District Guangdong Shenzhen518055 China

Non-overlapping codes are a set of codewords such that the prefix of each codeword is not a suffix of any codeword in the set, including itself. If the lengths of the codewords are variable, it is additionally required that every codeword is not contained in any other codeword as a subword. Let C(n, q) be the maximum size of q-ary fixed-length non-overlapping codes of length n. The upper bound on C(n, q) has been well studied. However, the nontrivial upper bound on the maximum size of variable-length non-overlapping codes of length at most n remains open. In this paper, by establishing a link between variable-length non-overlapping codes and fixed-length ones, we are able to show that the size of a q-ary variable-length non-overlapping code is upper bounded by C(n, q). Furthermore, we prove that the average length of the codewords in a q-ary variable-length non-overlapping codes is lower bounded by ⌈logq C˜⌉, and is asymptotically no shorter than n−2 as q approaches ∞, where C˜ denotes the cardinality of q-ary variable-length non-overlapping codes of length up to *** Codes 94A45, 94B25 © 2024, CC BY.

关键词： C (programming language)

来源：评论

学校读者我要写书评

暂无评论

CHEMFUZZ: Large Language Models-Assisted Fuzzing for Quantum Chemistry software Bug Detection

CHEMFUZZ: Large Language Models-Assisted Fuzzing for Quantum...

引用

IEEE International Conference on software Quality, Reliability and Security Companion (QRS-C)

作者： Feng Qiu Pu Ji Baojian Hua Yang Wang School of Software Engineering University of Science and Technology of China China Suzhou Institute for Advanced Research University of Science and Technology of China China Computer Science Department The Johns Hopkins University USA

Quantum chemistry software implements the first principle quantum computation and is indispensable in both scientific research and chemical industries. Any bugs in such software will lead to serious consequences, thus defeating its trustworthiness and reliability. However, bug detection techniques for such software have not been fully investigated. In this paper, to fill this gap, we propose a novel approach to fuzz quantum chemistry software with the aid of Large Language Models (LLMs). Our basic idea is utilize LLMs to mutate and generate syntactic and semantic valid input files from seed inputs, by proving valuable domain-specific knowledge of chemistry. With this basic idea, we have designed and implemented CHEMFuzz, a fully automatic fuzzing framework to fuzz quantum chemistry software for bugs. Our evaluation of CHEMFUZZ leverages popular LLMs including GPT3.5, Claude-2, and Bart as test oracles to generate parameters to mutate inputs and analyze computation results. CHEMFUZZ detected 40 unique bugs, which have been classified and reported to developers, with a code coverage of 17.4%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An n-MOSFET Layout with Multi-Finger Z Gate for Radiation Tolerance 5

An n-MOSFET Layout with Multi-Finger Z Gate for Radiation To...

引用

5th IEEE International Conference on Knowledge Innovation and Invention, ICKII 2022

作者： Shan, Chan Liu, Ying Chen, Huan Huo, Dong-Yang Xu, Tian-Rong Jimei University College of Ocean Information Engineering Xiamen China Xiamen Institute of Software Technology Department of Software Technology Xiamen China Minjiang University School of Journalism and Communication Fuzhou China Universiti Malaya Faculty of Computer Science and Information Technology Department of Information Systems Kuala Lumpur Malaysia Universiti Malaya Planning Faculty of Built Environment Department of Urban and Regional Kuala Lumpur Malaysia

ISBN: (数字)9781665479295

ISBN: (纸本)9781665479295

In this paper, a new n-MOSFET layout with multi-finger Z gate is proposed to reduce the total ionizing dose (TID) effect. In addition to the proposed layout, multi-finger single gate layout is also simulated using Sentaurus TCAD in 3D. Firstly, drain current - gate voltage curves of both layouts are compared in order to verify the radiation hardening ability of the proposed layout. Secondly, we compare TID responses by extracting threshold voltages and leakage currents from both layouts. Thirdly, the finger number is investigated for both layouts because it is an important factor for multi-finger layout. The results have indicated that the finger number significantly affects the multi-finger single gate layout, whereas the proposed multi-finger Z gate layout is relatively little affected by the finger number in radiation environment. © 2022 IEEE.

关键词： Drain current

来源：评论

学校读者我要写书评

暂无评论

E-VarM: Enhanced Variational Word Masks to Improve the Interpretability of Text Classification Models 29

E-VarM: Enhanced Variational Word Masks to Improve the Inter...

引用

29th International Conference on Computational Linguistics, COLING 2022

作者： Ge, Ling Hu, Chunming Ma, Guanghui Wu, Junshuang Chen, Junfan Liu, Jihong Zhang, Hong Qin, Wenyi Zhang, Richong School of Computer Science and Engineering Beihang University Beijing China College of Software Beihang University Beijing China National Computer Network Emergency Response Technical Team Beijing China Beijing Jinghang Research Institute of Computing and Communication Beijing China School of Mechanical Engineering and Automation Beihang University Beijing China

Enhancing the interpretability of text classification models can help increase the reliability of these models in real-world applications. Currently, most researchers focus on extracting task-specific words from inputs to improve the interpretability of the model. The competitive approaches exploit the Variational Information Bottleneck (VIB) to improve the performance of word masking at the word embedding layer to obtain task-specific words. However, these approaches ignore the multi-level semantics of the text, which can impair the interpretability of the model, and do not consider the risk of representation overlap caused by the VIB, which can impair the classification performance. In this paper, we propose an enhanced variational word masks approach, named E-VarM, to solve these two issues effectively. The E-VarM combines multi-level semantics from all hidden layers of the model to mask out task-irrelevant words and uses contrastive learning to readjust the distances between representations. Empirical studies on ten benchmark text classification datasets demonstrate that our approach outperforms the SOTA methods in simultaneously improving the interpretability and accuracy of the model. © 2022 Proceedings - International Conference on Computational Linguistics, COLING. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Robust Feature engineering Techniques for Designing Efficient Motor Imagery-Based BCI-systems

arXiv

引用

arXiv 2024年

作者： Gardezi, Syed Saim Jawed, Soyiba Khan, Mahnoor Bukhari, Muneeba Khan, Rizwan Ahmed Department of Biomedical Engineering Faculty of Engineering Salim Habib University Karachi Pakistan Department of Computer Software Engineering College of Electrical and Mechanical Engineering NUST Islamabad Pakistan Department of Physics Faculty of Applied Sciences Riphah International University Islamabad Pakistan Department of Biomedical Engineering Faculty of Engineering NED University Karachi Pakistan Department of Computer Science School of Mathematics and Computer Science Institute of Business Administration Karachi Pakistan

A multitude of individuals across the globe grapple with motor disabilities. Neural prosthetics utilizing Brain-computer Interface (BCI) technology exhibit promise for improving motor rehabilitation outcomes. The intricate nature of EEG data poses a significant hurdle for current BCI systems. Recently, a qualitative repository of EEG signals tied to both upper and lower limb execution of motor and motor imagery tasks has been unveiled. Despite this, the productivity of the Machine Learning (ML) Models that were trained on this dataset was alarmingly deficient, and the evaluation framework seemed insufficient. To enhance outcomes, robust feature engineering (signal processing) methodologies are implemented. A collection of time domain, frequency domain, and wavelet-derived features was obtained from 16-channel EEG signals, and the Maximum Relevance Minimum Redundancy (MRMR) approach was employed to identify the four most significant features. For classification K Nearest Neighbors, Support Vector Machine, Decision Tree, and Naïve Bayes models were implemented with these selected features, evaluating their effectiveness through metrics such as testing accuracy, precision, recall, and F1 Score. By leveraging SVM with a Gaussian Kernel, a remarkable maximum testing accuracy of 92.50% for motor activities and 95.48% for imagery activities is achieved. These results are notably more dependable and gratifying compared to the previous study, where the peak accuracy was recorded at 74.36%. This research work provides an in-depth analysis of the MI Limb EEG dataset and it will help in designing and developing simple, cost-effective and reliable BCI systems for neuro-rehabilitation. © 2024, CC BY-NC-ND.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Bayesian hypergame approach to equilibrium stability and robustness in moving target defense

Bayesian hypergame approach to equilibrium stability and rob...

引用

IEEE Conference on Decision and Control

作者： Hanzheng Zhang Zhaoyang Cheng Guanpu Chen Karl Henrik Johansson Shanghai Research Institute for Autonomous Intelligent Systems Tongji University Shanghai China Academy of Mathematics and System Science Chinese Academy of Science University of Chinese Academy of Science Beijing China Division of Decision and Control Systems School of Electrical Engineering and Computer Science KTH Royal Institute of Technology and the Digital Futures Stockholm Sweden

ISBN: (数字)9798350316339

ISBN: (纸本)9798350316346

We investigate the equilibrium stability and robustness in a class of moving target defense problems, in which players have both incomplete information and asymmetric cognition. We first establish a Bayesian Stackelberg game model for incomplete information and then employ a hypergame reformulation to address asymmetric cognition. With the core concept of the hyper Bayesian Nash equilibrium (HBNE), a condition for achieving both the strategic and cognitive stability in equilibria can be realized by solving linear equations. Moreover, to deal with players’ underlying perturbed knowledge, we study the equilibrium robustness by presenting a condition of robust HBNE under the given configuration. Experiments evaluate our theoretical results.

关键词： Games Nash equilibrium Stability analysis Robustness Mathematical models Cognition Bayes methods

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning for Unrelated Parallel Machine Scheduling with Release Dates, Setup Times, and Machine Eligibility 24

Reinforcement Learning for Unrelated Parallel Machine Schedu...

引用

Proceedings of the Winter Simulation Conference

作者： Sang-Hyun Cho Hyun-Jung Kim Lars Mönch Dept. of Industrial and Systems Engineering Korea Advanced Institute of Science and Technology Daejeon Republic of Korea Dept. of Mathematics and Computer Science University of Hagen Universitätsstraße 1 Hagen GERMANY

ISBN: (纸本)9798331534202

This paper presents a novel approach for solving unrelated parallel machine scheduling problems through reinforcement learning. Notably, we consider three main constraints: release date, machine eligibility, and sequence- and machine-dependent setup time to minimize total weighted tardiness. Our work presents a new graph representation for solving the problem and utilizes graph neural networks combined with reinforcement learning. Experimental results show that our proposed method outperforms traditional dispatching rules and an apparent tardiness cost-based algorithm. Furthermore, since we represent and solve the problem using graphs, our method can be used regardless of the number of jobs or machines once trained.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Wirelessly Actuated Rotation-free Magnetic Motor

Wirelessly Actuated Rotation-free Magnetic Motor

引用

IEEE/RSJ International Conference on Intelligent Robots and systems (IROS)

作者： Umur Ulas Harman Ahmed Hafez Cameron Duffield Zihan Zhao Luke Dixon Daniela Rus Shuhei Miyashita Automatic Control and Systems Engineering Department University of Sheffield Sheffield UK Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology Cambridge MA USA Insigneo Institute for in Silico Medicine

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

This paper addresses the challenge of actuating millimetre-sized motors, which are wirelessly driven by external magnetic fields. Traditional approaches, relying on rotating magnetic fields, often inadvertently cause the entire robot – especially if it is small and lightweight – to rotate, instead of a specified shaft in the motor. To overcome this issue, our study introduces a novel mechanism that leverages symmetrically configured magnetic motors to cancel out the torques, thus preventing unwanted rotation of the robot. This is achieved by utilizing a magnetic field along a single axis to induce rotational movement. The design features two millimetre-sized rotating magnets that interact to achieve a 90 ◦ rotation, complemented by an external magnetic field that accomplishes the remaining 270 ◦ , thus completing a full rotation. Furthermore, we demonstrate that applying a perpendicularly oriented magnetic field can inversely affect the motor’s rotation direction. A proof-of-concept experiment employing this mechanism successfully actuated a gripper in a water tank while it is free-floating, showcasing its potential for enhancing robotic applications at the sub-centimeter scale, where the small net torque of a miniature motor is essential.

关键词： Water Shafts Torque Gears Net zero Storage management Motors Permanent magnets Magnetic fields Grippers

来源：评论

学校读者我要写书评

暂无评论

FEAT: Towards Fast Environment-Adaptive Task Offloading and Power Allocation in MEC

FEAT: Towards Fast Environment-Adaptive Task Offloading and ...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE computer and Communications Societies

作者： Tao Ren Zheyuan Hu Hang He Jianwei Niu Xuefeng Liu Laboratory for Internet Software Technologies Institute of Software Chinese Academy of Sciences Beijing China State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Zhongguancun Laboratory Beijing China Zhengzhou University Research Institute of Industrial Technology School of Information Engineering Zhengzhou University Zhengzhou China

Mobile edge computing (MEC) has been proposed to provide mobile devices with both satisfactory computing resources and latency. Key issues in MEC include task offloading and power allocation (TOPA), for which deep reinforcement learning (DRL) is becoming a popular methodology. However, most DRL-based TOPA approaches are typically developed in a certain environment, lacking the adaptability to unseen environments. Motivated by this, this paper proposes a Fast Environment-Adaptive TOPA (FEAT) approach that could adapt to unseen environments with little fine-tuning. Specifically, we first split MEC states into the internal state and environmental state. Then, based on these two types of states, we develop two main components of FEAT — a group of internal state-dependent TOPA meta-policies and an environmental state-embedded steerer. Meta-policies learn TOPA skills within the internal state space (allowing reusing meta-policies in different environments), while the steerer learns to choose appropriate meta-policies according to embedded environmental states. When encountering an unseen environment with the same internal state space, FEAT only needs to fine-tune the steerer using the newly embedded environmental state with few internal state explorations. Extensive experimental results on simulation and testbed demonstrate that FEAT outperforms the state-of-the-art by more than 16.4% in terms of fine-tuning speeds.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：