JZTXT

efficient

SHARPNESS-AWARE MINIMIZATION FOR EFFICIENTLY IMPROVING GENERALIZATION论文阅读笔记

Intro 在训练集上最小化损失很可能导致泛化性低，因为当今模型的过参数化会导致training loss的landscape异常复杂且非凸，包含很多local/global minima，因此优化器的选择至关重要。loss landscape的几何性质（特别是minima的flatness）与泛化 ......
SHARPNESS-AWARE GENERALIZATION MINIMIZATION EFFICIENTLY SHARPNESS

LocPatcH An efficient long-read hybrid error correction algorithm based on local pHMM

该文档主要介绍了一种基于装配的方法和概率隐藏马尔科夫模型 (pHMM) 用于纠正长读序列的错误。文档详细描述了对酵母数据进行实验的结果、纠正方法的拓扑结构以及实验设置和数据集。 这种基于装配的纠正方法相对于直接纠正存在哪些优势？ pHMM 的拓扑结构是怎样的？ 在实验中使用了什么样的数据集？ 提示： ......
correction efficient algorithm long-read LocPatcH

UNetFormer: A UNet-like transformer for efficient semantic segmentation of remote sensing urban scene imagery

UNetFormer: A UNet-like transformer for efficient semantic segmentation of remote sensing urban scene imagery * Authors: [[Libo Wang]], [[Rui Li]], [[ ......
segmentation transformer UNetFormer UNet-like efficient

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network * Authors: [[Wenzhe Shi]], [[Jose Caballer ......
Super-Resolution Convolutional Resolution Real-Time Efficient

High-Efficiency Lossy Image Coding Through Adaptive Neighborhood Information Aggregation

目录简介创新点内容Entropy Coding Using Multistage Context Model模型结构残差邻域注意力块Residual Neighborhood Attention Block RNAB激活函数 高斯误差线性单元激活函数GELU并行解码 简介 创新点 Integrate ......
High-Efficiency Neighborhood Aggregation Information Efficiency

【HFSS】仿真时radiation efficiency（辐射效率）大于1

![image](https://img2023.cnblogs.com/blog/2603751/202311/2603751-20231121124817927-404911195.png) ![image](https://img2023.cnblogs.com/blog/2603751/20... ......
efficiency radiation 效率 HFSS

TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation

目录概TallRec代码 Bao K., Zhang J., Zhang Y., Wang W., Feng F. and He X. TALLRec: An effective and efficient tuning framework to align large language model ......
Recommendation Effective Efficient Framework Language

2023CVPR_Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring

一. Motivation 1. Transformer在解决全局表现很好，但是复杂度很高，主要体现在QK的乘积: (We note that the scaled dot-product attention computation is actually to estimate the corre ......
CVPR_Efficient Domain-based Transformers High-Quality Deblurring

Checkerboard Context Model for Efficient Learned Image Compression

目录AbstractIntroductionPreliminary 初步介绍Variational Image Compression with Hyperprior(超先验变分图像压缩)Autoregressive Context(自回归上下文模型)Parallel Context Modelin ......
Checkerboard Compression Efficient Context Learned

ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Coding

abstruct \(\quad\) 受能量压缩表现的启发，提出了不均匀通道情况自适应编码.结合不均匀分组模型和现有上下文模型，获得一种空间通道上下文自适应模型，来提高编码性能，而不影响其运行时间。 \(\quad\)这种模型支持预览解码和渐进解码。 introduction 学习图像压缩中最重要的 ......
Space-Channel Compression Contextual Efficient Adaptive

【CVPR2023】Efficient and Explicit Modelling of Image Hierarchies for Image Restoration

> 论文：https://readpaper.com/paper/4728855966703960065 代码：https://github.com/ofsoundof/GRL-Image-Restoration 这个论文的代码地址叫GRL，意思是 Global, Regional, Local 的 ......
Image Hierarchies Restoration Efficient Modelling

2023ACMMM_Mutual Information-driven Triple Interaction Network for Efficient Image Dehazing

一. Motivation 之前网络存在的缺点： 1. 使用的有限的频域信息 2. 不充足的信息交互 ： (1) 第一阶段的输出直接作为第二阶段的输入，忽略了中间特征从早期到后期的传播 (2) 在编码器解码器结构同尺度之间进行特征融合，忽略了阶段内和跨阶段的跨尺度信息交换 3. 严重的特征冗余：中间 ......
Information-driven ACMMM_Mutual Information Interaction Efficient

论文阅读：A Lightweight Knowledge Graph Embedding Framework for Efficient Inference and Storage

ABSTRACT 现存的KGE方法无法适用于大规模的图（由于存储和推理效率的限制） 作者提出了一种LightKG框架： 自动的推断出码本codebooks和码字codewords，为每个实体生成合适的embedding。 同时，框架中包含残差模块来实现码本的多样性，并且包含连续函数来近似的实现码字的 ......
Lightweight Knowledge Embedding Efficient Framework

Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables

郑重声明：原文参见标题，如有侵权，请联系作者，将会撤销发布！ Proceedings of the 36th International Conference on Machine Learning, PMLR 97:5331-5340, 2019 ......
Meta-Reinforcement Reinforcement Probabilistic Off-Policy Efficient

[VLDB 2012]Efficient Subgraph Matching on Billion Node Graphs

[VLDB 2012]Efficient Subgraph Matching on Billion Node Graphs 重点了解实现star-join的具体过程。 分解query和STwigs排序 文中把star叫做STwigs，每一个STwigs查询为\(q=(r, L)\)，其中r是跟节点标 ......
Efficient Subgraph Matching Billion Graphs

Proj CDeepFuzz Paper Reading: SparseProp: Efficient Sparse Backpropagation for Faster Training of Neural Networks

## Abstract 本文：SparseProp Github: https://github.com/IST-DASLab/sparseprop Task: a back-propagation algo for sparse training data, a fast vectorized i ......
Backpropagation SparseProp CDeepFuzz Efficient Networks

​MPDIoU: A Loss for Efficient and Accurate Bounding Box Regression

​MPDIoU: A Loss for Efficient and Accurate Bounding Box Regression MPDIoU:一个有效和准确的边界框损失回归函数 摘要 边界框回归(Bounding box regression, BBR)广泛应用于目标检测和实例分割，是目标定位 ......
Regression Efficient Accurate Bounding MPDIoU

Proj CDeepFuzz Paper Reading: Aries: Efficient Testing of Deep Neural Networks via Labeling-Free Accuracy Estimation

## Abstract 背景： 1. the de facto standard to assess the quality of DNNs in the industry is to check their performance (accuracy) on a collected set of ......
Labeling-Free Estimation CDeepFuzz Efficient Accuracy

[论文精读][计算生物][蛋白质预训练表示]Data-Efficient Protein 3D Geometric Pretraining via Refinement of Diffused Protein Structure Decoy

笔者正在调研市面上的蛋白表示方法，论文方法过于数理的部分会被抽象带过。 ## Basic Information: * Title: Data-Efficient Protein 3D Geometric Pretraining via Refinement of Diffused Protein St ......
Protein Data-Efficient 蛋白质 Pretraining Refinement

Efficient and Accurate Diagnostic Tool

Diagnostic tools play a crucial role in the automotive industry, allowing technicians to accurately identify and troubleshoot vehicle issues. Among th ......
Diagnostic Efficient Accurate Tool and

Classical Management: emphasized rationality and making organizations and workers as efficient as possible

Classical approach: First studies of management, which emphasized: * rationality * making organizations and workers as efficient as possible **Max Web ......
organizations rationality Management emphasized Classical

GLoRA：One-for-All: Generalized LoRA for Parameter-Efficient Fine-tuning

# GLoRA：One-for-All: Generalized LoRA for Parameter-Efficient Fine-tuning ## O、Abstract 本文在 LoRA 的基础上，提出一种广义 LoRA （GLoRA，Generalized LoRA）。与 LoRA 相比，G ......
Parameter-Efficient Fine-tuning One-for-All Generalized Efficient

[VLDBJ 2022]Privacy and efficiency guaranteed social subgraph matching

# Privacy and efficiency guaranteed social subgraph matching ## 动机 目标是在不影响查询处理的同时保护隐私 ## 其中的子图匹配算法PGP ![img](https://img2023.cnblogs.com/blog/2988279/ ......
efficiency guaranteed matching subgraph Privacy

Efficient GPU-Accelerated Subgraph Matching

# Efficient GPU-Accelerated Subgraph Matching ## 总结 核心在利用GPU并行计算，为此设计了更适合GPU查询的数据结构，并混合BFS-DFS（先广度过滤再深度匹配）实现更好的时空复杂度 ## 动机 现有的算法都是先过滤再枚举。常规的CPU算法一次只能计 ......
GPU-Accelerated Accelerated Efficient Subgraph Matching

April 2023-Memory-efficient Reinforcement Learning with Value-based Knowledge Consolidation

本文基于深度q网络算法提出了记忆高效的强化学习算法来缓解这一问题。通过将目标q网络中的知识整合Knowledge Consolidation到当前q网络中，所提算法减少了遗忘并保持了较高的样本效率。 ......
Memory-efficient Consolidation Reinforcement Value-based Knowledge

Efficient Correction of Single InsertionlDeletion and Multi-Substitution Errors

Efficient Correction of Single InsertionlDeletion and Multi-Substitution ErrorsG. J. Han, Y. L. Guan, K. Cai, K. S. Chan, and L. J. Kong A!JshYlc�A tw ......
InsertionlDeletion Multi-Substitution Substitution Correction Efficient

Efficient Graph Generation with Graph Recurrent Attention Networks

[TOC] > [Liao R., Li Y., Song Y., Wang S., Nash C., Hamilton W. L., Duvenaud D., Urtasun R. and Zemel R. NIPS, 2019.](http://arxiv.org/abs/1910.00760) ......
Graph Generation Efficient Attention Recurrent

Professional C++阅读笔记 chapter 29 Writing Efficient C++

# chapter 29 Writing Efficient C++ 1. 作者建议将所有class function包括析构函数 但是除了构造函数，都设为virtual的，因为virtual function的开销非常小 2. design 和 algrithm 比语言层面的优化重要太多 3. 在 ......
Professional Efficient chapter Writing 笔记

June 2021-Continuous Transition: Improving Sample Efficiency for Continuous Control Problems via MixUp

本文建议通过对连续transition进行线性插值来合成新的transition用于训练。为了保持构建的transition的真实性，还开发了一个鉴别器来自动指导构建过程 ......
Continuous Efficiency Transition Improving Problems

May 2022-Neighborhood Mixup Experience Replay: Local Convex Interpolation for Improved Sample Efficiency in Continuous Control Tasks

提出了邻域混合经验回放(NMER)，一种基于几何的回放缓冲区，用状态-动作空间中最近邻的transition进行插值。NMER仅通过混合transition与邻近状态-动作特征来保持trnaistion流形的局部线性近似。 ......
Interpolation Neighborhood Experience Continuous Efficiency
共39篇  :1/2页 首页上一页1下一页尾页