检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Lv, Weijie Xia, Xuan Huang, Sheng-Jun Nanjing University of Aeronautics and Astronautics Nanjing China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China

Large language models (LLMs) have shown great potential in code-related tasks, yet open-source models lag behind their closed-source counterparts. To bridge this performance gap, existing methods generate vast amounts of synthetic data for fine-tuning, leading to inefficiencies in training. Motivated by the need for more effective and efficient training, we propose the Code Adaptive Compute-efficient Tuning (CodeACT) framework. CodeACT introduces the Complexity and Diversity Aware Sampling (CDAS) method to select high-quality training data based on complexity and diversity, and the Dynamic Pack padding strategy to reduce computational resource usage by minimizing padding tokens during training. Experimental results demonstrate that CodeACT-DeepSeek-Coder-6.7B, fine-tuned on only 40% of the EVOL-Instruct data, achieves an 8.6% performance increase on HumanEval, reduces training time by 78%, and decreases peak GPU memory usage by 27%. These findings underscore CodeACT’s ability to enhance the performance and efficiency of open-source models. By optimizing both the data selection and training processes, CodeACT offers a comprehensive approach to improving the capabilities of open-source LLMs while significantly reducing computational requirements, addressing the dual challenges of data quality and training efficiency, and paving the way for more resource-efficient and performant models. The code is available at Kyle-Lyu/CodeACT. Copyright © 2024, The Authors. All rights reserved.

关键词： Data quality

来源：评论

学校读者我要写书评

暂无评论

Multi-kernel-size Convolutional Supervised Autoencoders for Tactile Gesture Recognition 12

Multi-kernel-size Convolutional Supervised Autoencoders for ...

引用

12th International Conference on CYBER Technology in Automation, Control, and Intelligent Systems, CYBER 2022

作者： Ye, Chaoxiang Li, Xiaoyu Huang, Binhua Su, Yuanzhe Mi, Tingting Zhou, Zhenning Yi, Zhengkun Wu, Xinyu Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen Institute of Advanced Technology Chinese Academy of Sciences China Siat Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518055 China

ISBN: (数字)9781665472678

ISBN: (纸本)9781665472678

With the development of human-computer interaction, tactile gesture recognition has been widely used in our life. To solve the problem of overfitting on the sample-limited tactile datasets, we apply the Supervised Autoencoders (SAE) to improve the generalization performance. Moreover, based on the SAE, we propose the Multi-kernel-size Convolutional Supervised Autoencoders (MCSAE) to further improve the generalization performance on the limited dataset, which provides models with more structure of receptive fields and enhances the feature extraction ability of SAE. In comparison with other state-of-the-art (SOTA) models, the SAE we apply has higher gesture recognition accuracy and MCSAE can further improve the generalization performance of SAE on the sample-limited publicly available dataset. © 2022 IEEE.

关键词： Gesture recognition

来源：评论

学校读者我要写书评

暂无评论

Enhancing Visual Continual Learning with Language-Guided Supervision

arXiv

引用

arXiv 2024年

作者： Ni, Bolin Zhao, Hongbo Zhang, Chenghao Hu, Ke Meng, Gaofeng Zhang, Zhaoxiang Xiang, Shiming State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Centre for Artificial Intelligence and Robotics HK Institute of Science & Innovation Chinese Academy of Sciences China

Continual learning (CL) aims to empower models to learn new tasks without forgetting previously acquired knowledge. Most prior works concentrate on the techniques of architectures, replay data, regularization, etc. However, the category name of each class is largely neglected. Existing methods commonly utilize the one-hot labels and randomly initialize the classifier head. We argue that the scarce semantic information conveyed by the one-hot labels hampers the effective knowledge transfer across tasks. In this paper, we revisit the role of the classifier head within the CL paradigm and replace the classifier with semantic knowledge from pretrained language models (PLMs). Specifically, we use PLMs to generate semantic targets for each class, which are frozen and serve as supervision signals during training. Such targets fully consider the semantic correlation between all classes across tasks. Empirical studies show that our approach mitigates forgetting by alleviating representation drifting and facilitating knowledge transfer across tasks. The proposed method is simple to implement and can seamlessly be plugged into existing methods with negligible adjustments. Extensive experiments based on eleven mainstream baselines demonstrate the effectiveness and generalizability of our approach to various protocols. For example, under the class-incremental learning setting on ImageNet-100, our method significantly improves the Top-1 accuracy by 3.2% to 6.1% while reducing the forgetting rate by 2.6% to 13.1%. © 2024, CC BY.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Effectively Detecting Loop Closures using Point Cloud Density Maps

Effectively Detecting Loop Closures using Point Cloud Densit...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Saurabh Gupta Tiziano Guadagnino Benedikt Mersch Ignacio Vizzo Cyrill Stachniss Center for Robotics University of Bonn Germany Dexory UK Lamarr Institute for Machine Learning and Artificial Intelligence Germany

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

The ability to detect loop closures plays an essential role in any SLAM system. Loop closures allow correcting the drifting pose estimates from a sensor odometry pipeline. In this paper, we address the problem of effectively detecting loop closures in LiDAR SLAM systems in various environments with longer lengths of sequences and agnostic of the scanning pattern of the sensor. While many approaches for loop closures using 3D LiDAR sensors rely on individual scans, we propose the usage of local maps generated from locally consistent odometry estimates. Several recent approaches compute the maximum elevation map on a bird’s eye view projection of point clouds to compute feature descriptors. In contrast, we use a density image bird’s eye view representation, which is robust to viewpoint changes. The utilization of dense local maps allows us to reduce the complexity of features describing these maps, as well as the size of the database required to store these features over a long sequence. This yields a real-time application of our approach for a typical robotic 3D LiDAR sensor. We perform extensive experiments to evaluate our approach against other state-of-the-art approaches and show the benefits of our proposed approach.

关键词： Point cloud compression Simultaneous localization and mapping Laser radar Three-dimensional displays Image coding Pipelines Image representation

来源：评论

学校读者我要写书评

暂无评论

A Meta-learning System for Automatic Objects Tracking 4

A Meta-learning System for Automatic Objects Tracking

引用

4th International Conference on Applied Machine Learning, ICAML 2022

作者： Li, Qiaolin Wang, Wenfeng Zhang, Jingjing Research Institute of Intelligent Engineering and Data Applications Shanghai Institute of Technology Shanghai China Research Center of Artificial Intelligence and Robotics Interscience Institute of Management and Technology Bhubaneswar India

ISBN: (纸本)9781665462655

Recent years, reinforcement learning has been developed in many area and achieve high performance in many application. However, for huge amount data and training will make hard leaning for the agent of reinforcement learning, in the project, the agent of DDQN Network was integrated with LSTM meta learner for learning the game called Cartpole and LunarLander. It get a good performance for large train. © 2022 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Light-resistant target detection improvement algorithm for overexposed environments

引用

Transportation Safety and Environment 2025年第1期7卷 147-157页

作者： Chen Zheng Zhiwei Li Hui Liu Sha Huang Yanjia Zhao School of Rail Transportation Wuyi UniversityJiangmen 529020China Key Laboratory of Intelligent Operation Maintenance and Emergency Management for Rail Transit of JiangmenWuyi UniversityJiangmen 529020China Institute of Artificial Intelligence&Robotics(IAIR) Key Laboratory of Traffic Safety on Track of Ministry of EducationSchool of Traffic and Transportation EngineeringCentral South UniversityChangsha 410075China

In strong light environments,images often appear overexposed,which seriously impacts the accuracy of target *** existing research,however,requires additional modules to assist in detection,which affects the timeliness of the detection *** address the issues of reduced target detection accuracy and timeliness in overexposed environments,this paper proposes a realtime anti-light target detection improvement algorithm based on you-only-look-once v8n(YOLO v8n),focusing on enhancing the model’s ability to extract features from overexposed images without the need for additional ***,online overexposure enhancement technology is integrated into model training to simulate overexposed images produced in overexposed environments,enhancing the model’s robustness in detecting overexposed *** convolution networks v2 is used to improve the cross-stage partial bottleneck with two convolutions layer,addressing the issue of traditional convolution’s poor feature extraction performance for overexposed images,thereby aiding the model in capturing targets with weakened or missing features and enhancing the model’s ability to construct the geometric shape of ***,large separable kernel attention is introduced to enhance the spatial pyramid pooling fast layer,strengthening the model’s overall connectivity for targets with missing ***,distance intersection over union is utilized to optimize the detection accuracy of overlapping targets in overexposed *** experimental results show that,compared to the original model,the mAP50 and mAP50–95 of the model designed in this paper are improved by 23.2%and 15.7%,respectively,and the model size only increases by 0.3 *** improving detection accuracy,the lightweight requirements for actual deployment are also met.

关键词： light-resistant target detection you only look once(YOLO)v8 image overexposure online data augmentation

来源：评论

学校读者我要写书评

暂无评论

Modeling and Control of Intrinsically Elasticity Coupled Soft-Rigid Robots

Modeling and Control of Intrinsically Elasticity Coupled Sof...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Zach J. Patterson Cosimo Della Santina Daniela Rus Computer Science and Artificial Intelligence Laboratory MIT Department of Cognitive Robotics Delft University of Technology Institute of Robotics and Mechatronics German Aerospace Center (DLR)

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

While much work has been done recently in the realm of model-based control of soft robots and soft-rigid hybrids, most works examine robots that have an inherently serial structure. While these systems have been prevalent in the literature, there is an increasing trend toward designing soft-rigid hybrids with intrinsically coupled elasticity between various degrees of freedom. In this work, we seek to address the issues of modeling and controlling such structures, particularly when underactuated. We introduce several simple models for elastic coupling, typical of those seen in these systems. We then propose a controller that compensates for the elasticity, and we prove its stability with Lyapunov methods without relying on the elastic dominance assumption. This controller is applicable to the general class of underactuated soft robots. After evaluating the controller in simulated cases, we then develop a simple hardware platform to evaluate both the models and the controller. Finally, using the hardware, we demonstrate a novel use case for underactuated, elastically coupled systems in "sensorless" force control.

关键词： Couplings Elasticity Soft robotics Robot sensing systems Market research Hardware Stability analysis

来源：评论

学校读者我要写书评

暂无评论

Quantized Distillation: Optimizing Driver Activity Recognition Models for Resource-Constrained Environments

arXiv

引用

arXiv 2023年

作者： Tanama, Calvin Peng, Kunyu Marinov, Zdravko Stiefelhagen, Rainer Roitberg, Alina Institute for Anthropomatics and Robotics Karlsruhe Institute of Technology Germany Institute for Artificial Intelligence University of Stuttgart Germany

Deep learning-based models are at the top of most driver observation benchmarks due to their remarkable accuracies but come with a high computational cost, while the resources are often limited in real-world driving scenarios. This paper presents a lightweight framework for resource-efficient driver activity recognition. We enhance 3D MobileNet, a speed-optimized neural architecture for video classification, with two paradigms for improving the trade-off between model accuracy and computational efficiency: knowledge distillation and model quantization. Knowledge distillation prevents large drops in accuracy when reducing the model size by harvesting knowledge from a large teacher model (I3D) via soft labels instead of using the original ground truth. Quantization further drastically reduces the memory and computation requirements by representing the model weights and activations using lower precision integers. Extensive experiments on a public dataset for in-vehicle monitoring during autonomous driving show that our proposed framework leads to an 3–fold reduction in model size and 1.4–fold improvement in inference time compared to an already speed-optimized architecture. Our code is available at https://***/ calvintanama/qd-driver-activity-reco. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Towards Extreme Image Compression with Latent Feature Guidance and Diffusion Prior

arXiv

引用

arXiv 2024年

作者： Li, Zhiyuan Zhou, Yanhui Wei, Hao Ge, Chenyang Jiang, Jingwen National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an710049 China

Image compression at extremely low bitrates (below 0.1 bits per pixel (bpp)) is a significant challenge due to substantial information loss. In this work, we propose a novel two-stage extreme image compression framework that exploits the powerful generative capability of pre-trained diffusion models to achieve realistic image reconstruction at extremely low bitrates. In the first stage, we treat the latent representation of images in the diffusion space as guidance, employing a VAE-based compression approach to compress images and initially decode the compressed information into content variables. The second stage leverages pre-trained stable diffusion to reconstruct images under the guidance of content variables. Specifically, we introduce a small control module to inject content information while keeping the stable diffusion model fixed to maintain its generative capability. Furthermore, we design a space alignment loss to force the content variables to align with the diffusion space and provide the necessary constraints for optimization. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art approaches in terms of visual performance at extremely low bitrates. The source code and trained models are available at https://***/huai-chang/DiffEIC. Copyright © 2024, The Authors. All rights reserved.

关键词： Image compression

来源：评论

学校读者我要写书评

暂无评论

Polarization-controlled dual resonant lattice Kerker effects

引用

Nano Research 2023年第2期16卷 3195-3200页

作者： Lei Xiong Xueqian Zhao Xiang Du Shuai Chen Yuanfu Lu Hongwei Ding Guangyuan Li School of Information Science and Engineering Yunnan UniversityKunming 650500China CAS Key Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Institute of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China School of Computer Electronic and InformationGuangxi UniversityNanning 530004China College of Physics and Electronic Technology Anhui Normal UniversityWuhu 241000China Tianjin H-Chip Technology Group Corporation Tianjin 300467China SIAT Branch Shenzhen Institute of Artificial Intelligence and Robotics for SocietyShenzhen 518055China

The generalized Kerker effects have attracted increasing interests in recent years due to their abilities to manipulate the far-field properties of ***,the dual-polarized generalized Kerker effect enabling different tailoring of orthogonally-polarized electromagnetic waves has not yet been ***,we demonstrate polarization-controlled dual resonant lattice Kerker effects in periodic silicon *** varying the incident angle,the electric dipole and magnetic dipole surface lattice resonances can spectrally overlap,causing zero reflectance and unitary transmittance,i.e.,the resonant lattice Kerker *** incident angle for achieving this effect can be tuned differently for s-and p-polarizations over large regions by varying the nanodisk size or the lattice *** proposed dual-polarized resonant lattice Kerker effects open up avenues for polarization-controlled manipulation of the phase and wavefront of light with metasurfaces.

关键词： Kerker effect surface lattice resonances dual-polarized silicon metasurface

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：