deep-learning-essay's People
deep-learning-essay's Issues
Understanding and Robustifying Differentiable Architecture Search
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
One sentence
通过学习一个模型无关的元学习器,使得能快速adapt到相似的任务上。理论上能在任意的模型上套上这个元学习器。
Paper
https://arxiv.org/abs/1703.03400
Code
https://github.com/cbfinn/maml
Motivation
Novelties & Key contribution
Method
Experiments
Rethinking
Reference
https://zhuanlan.zhihu.com/p/57864886
https://towardsdatascience.com/paper-repro-deep-metalearning-using-maml-and-reptile-fd1df1cc81b0
Single Path One-Shot Neural Architecture Search with Uniform Sampling
旷视zhangxiangyu团队。
One-Shot表示超网训练完以后,直接通过某种方式(如采样),生成某种子网,而生成的子网,不做任何fine-tuning,直接用来评估子网的推理精度,这可以大大加快结构搜索的速度。而这里用超网参数赋给子网,以这个结果来表示最终子网训练收敛的结果,这个代理的好坏直接影响了One-Shot方法的效果。
Single Path则是训练超网的方法。训练超网时,不是类似DARTS这样,所有的结构以一定的权重带入网络,而是每次,只随机取一条激活的路径。
ADVERSARIAL AUTOAUGMENT
基于GAN实现的自动数据增强
SM-NAS: Structural-to-Modular Neural Architecture Search for Object Detection
One sentence
华为诺亚的文章,分为两个stage,用coarse-to-grain的方式来搜:
1)Stage-One:搜分辨率、backbone结构(ResNet18 or ResNet50 or ResNext or MobileNetV2)、Feature Fusion(With or without)RPN(with or without)、RCNN Head;Stage-One使用ImageNet预训练模型进行fine-tuning
2)Stage-Two:搜Backbone,直接在COCO上train from scratch
Paper
https://arxiv.org/abs/1911.09929
https://arxiv.org/pdf/1911.09929.pdf
Code
Motivation
Novelties & Key contribution
Method
Experiments
Rethinking
Reference
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
BlockQNN: Efficient Block-wise Neural Network Architecture Generation
Why Gradient Clipping Accelerates Training: A Theoretical Justification for Adaptivity
Understanding Architectures Learnt by Cell-based Neural Architecture Search
Lineage Stash: Fault Tolerance Off the Critical Path
Res2Net: A New Multi-scale Backbone Architecture
Random Search and Reproducibility for Neural Architecture Search
AutoML-Zero: Evolving Machine Learning Algorithms From Scratch
MixMatch: A Holistic Approach to Semi-Supervised Learning
Single-Path Mobile AutoML: Efficient ConvNet Design and NAS Hyperparameter Optimization
可以搜索channel,feature map size,
A Closer Look at Deep Policy Gradients
Understanding and Robustifying Differentiable Architecture Search
MTL-NAS: Task-Agnostic Neural Architecture Search towards General-Purpose Multi-Task Learning
Neural Optimizer Search with Reinforcement Learning
ONCE FOR ALL: TRAIN ONE NETWORK AND SPECIALIZE IT FOR EFFICIENT DEPLOYMENT
训练一个网络,根据不同的时延需求,从网络中采样不同的子网络来使用。
https://arxiv.org/abs/1812.00332
Multi-objective Neural Architecture Search via Predictive Network Performance Optimization
SEARCHING FOR ACTIVATION FUNCTIONS
Bounding Box Regression with Uncertainty for Accurate Object Detection
MobileNetV2: Inverted Residuals and Linear Bottlenecks
https://arxiv.org/pdf/1801.04381.pdf
Linear Bottlenecks:
Inverted Residual:
ResNet中的Residual Block,主分支由一个1x1,一个3x3,一个1x1三个卷积结构组成,其中,第一个1x1会将channel数减小到原来的1/4。而Inverted Residual则正好相反,第一个1x1会将channel数升高。
这样的作用是
Identity Mappings in Deep Residual Networks
ResNet的第二篇文章
Deep Residual Learning for Image Recognition
Accelerating Neural Architecture Search using Performance Prediction
R-FCN: Object Detection via Region-based Fully Convolutional Networks
https://arxiv.org/pdf/1605.06409.pdf
Faster-RCNN的改进
IRLAS: Inverse Reinforcement Learning for Architecture Search
Enhancing Generalization of First-Order Meta-Learning
Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression
original paper:
https://arxiv.org/abs/1902.09630
chinese :
https://zhuanlan.zhihu.com/p/57992040
Regularized Evolution for Image Classifier Architecture Search
One-Sentence
Aging策略的遗传算法作为sampler来搜结构。
Paper
https://arxiv.org/pdf/1802.01548.pdf
Code
Q.1 为什么叫Regularized?
因为在遗传算法中用了Aging的策略,也就是说,淘汰最老,而不是淘汰最差。
其中的一个关键点是:为什么每次去除的是最“老”的模型,而不是性能最差的模型?原因是作者认为如果一个模型效果足够好,那么他有很大概率在他被淘汰之前已经留下了自己的后代。如果每次淘汰的是最差的样本的话,那么队列中的(现存的)样本很有可能是来自一个共同祖先,我们得到的结构很容易失去多样性。这种在遗传学上就叫做近亲繁殖。
作者:无影
链接:https://zhuanlan.zhihu.com/p/81126873
来源:知乎
著作权归作者所有。商业转载请联系作者获得授权,非商业转载请注明出处。
Learning Spatial Fusion for Single-Shot Object Detection
NAS-Bench-101: Towards Reproducible Neural Architecture Search
Efficient Neural Architecture Search via Proximal Iterations
Google Vizier: A Service for Black-Box Optimization
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
https://arxiv.org/pdf/1704.04861.pdf
提出了Depthwise separable convolution(深度可分离卷积),将传统卷积分为了两个部分:depthwise convolution 和 pointwise convolution,大大减少卷积操作的参数量,减少为大约1 / K ** 2,K是卷积核的大小 。
但实际应用中效果并不是很理想,尤其是在GPU推理时性能极差,在移动端推理比较有用。
Evaluating The Search Phase of Neural Architecture Search
https://openreview.net/forum?id=H1loF2NFwr
ICLR 2020 accepted
REVISITING BATCH NORMALIZATION FOR PRACTICAL DOMAIN ADAPTATION
Bag of Freebies for Training Object Detection Neural Networks
ADVERSARIAL AUTOAUGMENT
A Simple Framework for Contrastive Learning of Visual Representations
ACCELERATING FRAMEWORK FOR SIMULTANEOUS OPTIMIZATION OF MODEL ARCHITECTURES AND TRAINING HYPERPARAMETERS
BETANAS: BalancEd TrAining and selective drop for Neural Architecture Search
Understanding and Simplifying One-Shot Architecture Search
One-Shot方法的鼻祖文章
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
概述
把NAS搜索看作是Path Pruning的问题。
可以看作是SinglePath One-Shot的始祖,将网络二值化,来解决DARTS中显存占用高的问题。
文章:https://arxiv.org/abs/1812.00332
代码:https://github.com/mit-han-lab/ProxylessNAS/tree/master/search
作者解读:https://www.zhihu.com/question/296404213/answer/547163236
NAS evaluation is frustratingly hard
RON: Reverse Connection with Objectness Prior Networks for Object Detection
https://arxiv.org/pdf/1707.01691.pdf
Faster-RCNN的改进
Understanding Why Neural Networks Generalize Well Through GSNR of Parameters
NAS-BENCH-201: EXTENDING THE SCOPE OF REPRODUCIBLE NEURAL ARCHITECTURE SEARCH
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.