关于深度学习理论和架构的最新综述-附录

附录 A
A.1. Frameworks
A.2. SDKs
- A.3. Benchmark Datasets
- - A.3.1. Image Classification or Detection or Segmentation
  - A.3.2. Text Classification
  - A.3.3. Language Modeling
  - A.3.4. Image Captioning
  - A.3.5. Machine Translation
  - A.3.6. Question Answering
  - A.3.7. Speech Recognition
  - A.3.8. Document Summarization
  - A.3.9. Sentiment Analysis:
  - A.3.10. Hyperspectral Image Analysis
- A.4. Journals and Conferences
- - A.4.1. Conferences
  - A.4.2. Journal
  - A.4.3. Tutorials on Deep Learning
  - A.4.4. Books on Deep Learning

附录 A

大多数时候，人们使用不同的深度学习框架和标准开发套件（SDK）来实现深度学习方法，如下所示：

A.1. Frameworks

Tensorflow: https://www.tensorflow.org/
Caffe: http://caffe.berkeleyvision.org/
KERAS: https://keras.io/
Theano: http://deeplearning.net/software/theano/
Torch: http://torch.ch/
PyTorch: http://pytorch.org/
Lasagne: https://lasagne.readthedocs.io/en/latest/
DL4J (DeepLearning4J): https://deeplearning4j.org/
Chainer: http://chainer.org/
DIGITS: https://developer.nvidia.com/digits
CNTK (Microsoft): https://github.com/Microsoft/CNTK
MatConvNet: http://www.vlfeat.org/matconvnet/
MINERVA: https://github.com/dmlc/minerva
MXNET: https://github.com/dmlc/mxnet
OpenDeep: http://www.opendeep.org/
PuRine: https://github.com/purine/purine2
PyLerarn2: http://deeplearning.net/software/pylearn2/
TensorLayer: https://github.com/zsdonghao/tensorlayer
LBANN: https://github.com/LLNL/lbann

A.2. SDKs

cuDNN: https://developer.nvidia.com/cudnn
TensorRT: https://developer.nvidia.com/tensorrt
DeepStreamSDK: https://developer.nvidia.com/deepstream-sdk
cuBLAS: https://developer.nvidia.com/cublas
cuSPARSE: http://docs.nvidia.com/cuda/cusparse/
NCCL: https://devblogs.nvidia.com/parallelforall/fast-multi-gpu-collectives-nccl/

A.3. Benchmark Datasets

以下是经常用于评估不同应用领域深度学习方法的基准数据集列表：

A.3.1. Image Classification or Detection or Segmentation

数据集列表用于图像处理和计算机视觉领域：

MNIST: http://yann.lecun.com/exdb/mnist/
CIFAR 10/100: https://www.cs.toronto.edu/~{}kriz/cifar.html
SVHN/ SVHN2: http://ufldl.stanford.edu/housenumbers/
CalTech 101/256: http://www.vision.caltech.edu/Image_Datasets/Caltech101/
STL-10: https://cs.stanford.edu/~{}acoates/stl10/
NORB: http://www.cs.nyu.edu/~{}ylclab/data/norb-v1.0/
SUN-dataset: http://groups.csail.mit.edu/vision/SUN/
ImageNet: http://www.image-net.org/
National Data Science Bowl Competition: http://www.datasciencebowl.com/
COIL 20/100: http://www.cs.columbia.edu/CAVE/software/softlib/coil-20.php
MS COCO DATASET: http://mscoco.org/
MIT-67 scene dataset: http://web.mit.edu/torralba/www/indoor.html
Caltech-UCSD Birds-200 dataset: http://www.vision.caltech.edu/visipedia/CUB-200-2011.html
Pascal VOC 2007 dataset: http://host.robots.ox.ac.uk/pascal/VOC/voc2007/
H3D Human Attributes dataset: https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/shape/poselets/
Face recognition dataset: http://vis-www.cs.umass.edu/lfw/
For more data-set visit: https://www.kaggle.com/
http://homepages.inf.ed.ac.uk/rbf/CVonline/Imagedbase.htm
Recently Introduced Datasets in Sept. 2016:
Google Open Images (~9M images)—https://github.com/openimages/dataset
Youtube-8M (8M videos: https://research.google.com/youtube8m/

A.3.2. Text Classification

Reuters-21578 Text Categorization Collection: http://kdd.ics.uci.edu/databases/reuters21578/reuters21578.html
Sentiment analysis from Stanford: http://ai.stanford.edu/~{}amaas/data/sentiment/
Movie sentiment analysis from Cornel: http://www.cs.cornell.edu/people/pabo/movie-review-data/

A.3.3. Language Modeling

Free eBooks: https://www.gutenberg.org/
Brown and stanford corpus on present americal english: https://en.wikipedia.org/wiki/Brown_ Corpus
Google 1Billion word corpus: https://github.com/ciprian-chelba/1-billion-word-languagemodeling-benchmark

A.3.4. Image Captioning

Flickr-8k: http://nlp.cs.illinois.edu/HockenmaierGroup/8k-pictures.html

Flickr-30k
Common Objects in Context (COCO): http://cocodataset.org/#overview, http://sidgan.me/ technical/2016/01/09/Exploring-Datasets

A.3.5. Machine Translation

Pairs of sentences in English and French: https://www.isi.edu/natural-language/download/ hansard/
European Parliament Proceedings parallel Corpus 196-2011: http://www.statmt.org/europarl/
The statistics for machine translation: http://www.statmt.org/

A.3.6. Question Answering

Stanford Question Answering Dataset (SQuAD): https://rajpurkar.github.io/SQuAD-explorer/ Dataset from DeepMind: https://github.com/deepmind/rc-data
Amazon dataset: http://jmcauley.ucsd.edu/data/amazon/qa/, http://trec.nist.gov/data/ qamain…, http://www.ark.cs.cmu.edu/QA-data/, http://webscope.sandbox.yahoo.co…, http://blog.stackoverflow.com/20…

A.3.7. Speech Recognition

TIMIT: https://catalog.ldc.upenn.edu/LDC93S1
Voxforge: http://voxforge.org/
Open Speech and Language Resources: http://www.openslr.org/12/

A.3.8. Document Summarization

https://archive.ics.uci.edu/ml/datasets/Legal+Case+Reports
http://www-nlpir.nist.gov/related_projects/tipster_summac/cmp_lg.html
https://catalog.ldc.upenn.edu/LDC2002T31

A.3.9. Sentiment Analysis:

IMDB dataset: http://www.imdb.com/

A.3.10. Hyperspectral Image Analysis

http://www.ehu.eus/ccwintco/index.php/Hyperspectral_Remote_Sensing_Scenes
https://engineering.purdue.edu/~{}biehl/MultiSpec/hyperspectral.html
http://www2.isprs.org/commissions/comm3/wg4/HyRANK.html
此外，在数据编程中还有另一种替代解决方案，使用弱监督策略或域启发法作为标记函数标记数据子集，即使它们有噪声并且可能与样本冲突 ¹.

A.4. Journals and Conferences

一般而言，研究人员在ArXiv上发布他们的主要研究版本(https://arxiv.org/ ).大多数会议都接受了关于深度学习及其相关领域的论文。热门会议如下：

A.4.1. Conferences

Neural Information Processing System (NIPS)
International Conference on Learning Representation (ICLR): What are you doing for Deep Learning?
International Conference on Machine Learning (ICML)
Computer Vision and Pattern Recognition (CVPR): What are you doing with Deep Learning?
International Conference on Computer Vision (ICCV)
European Conference on Computer Vision (ECCV)
British Machine Vision Conference (BMVC)

A.4.2. Journal

Journal of Machine Learning Research (JMLR)
IEEE Transaction of Neural Network and Learning System (ITNNLS)
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
Computer Vision and Image Understanding (CVIU)
Pattern Recognition Letter
Neural Computing and Application
International Journal of Computer Vision
IEEE Transactions on Image Processing
IEEE Computational Intelligence Magazine
Proceedings of IEEE
IEEE Signal Processing Magazine
Neural Processing Letter
Pattern Recognition
Neural Networks
ISPPRS Journal of Photogrammetry and Remote Sensing

A.4.3. Tutorials on Deep Learning

http://deeplearning.net/tutorial/
http://deeplearning.stanford.edu/tutorial/
http://deeplearning.net/tutorial/deeplearning.pdf
Courses on Reinforcement Learning: http://rll.berkeley.edu/deeprlcourse/

A.4.4. Books on Deep Learning

https://github.com/HFTrader/DeepLearningBook
https://github.com/janishar/mit-deep-learning-book-pdf
http://www.deeplearningbook.org/

↩︎

关于深度学习理论和架构的最新综述(附录)相关推荐

关于深度学习理论和架构的最新综述(part3)
关于深度学习理论和架构的最新综述-part3 递归神经网络(RNN) 介绍长短期记忆(LSTM) 门控循环单元(GRU) 卷积LSTM(ConvLSTM) RNN架构的变体及其应用基于注意力的RN ...
深度学习助力网络科学：基于深度学习的社区发现最新综述
来源:AMiner科技论文题目: A Comprehensive Survey on Community Detection with Deep Learning 论文网址: https://arx ...
上海交大华为：“非完全监督下基于深度学习的图像分割方法”最新综述
上海交通大学人工智能研究院杨小康.沈为团队联合华为田奇团队共同发布了非完全监督(即标签有限)下的图像分割方法最新综述"A Survey on Label-efficient Deep Seg ...
深度学习理论与架构最新进展综述论文，66页pdf，333篇参考文献
[导读]本文章从深度神经网络(DNN)入手,对深度学习(DL)领域的研究进展进行了简要的综述.内容包括:卷积神经网络(CNN).循环神经网络(RNN).长时记忆(LSTM)和门控递归单元(GRU).自 ...
深度强化学习探索算法最新综述，近200篇文献揭示挑战和未来方向
©作者 | 杨天培.汤宏垚等来源 | 机器之心强化学习是在与环境交互过程中不断学习的,⽽交互中获得的数据质量很⼤程度上决定了智能体能够学习到的策略的⽔平.因此,如何引导智能体探索成为强化学习领域研 ...
深度学习图像超分辨率最新综述：从模型到应用
点击我爱计算机视觉标星,更快获取CVML新技术今日arXiv新上论文<Deep Learning for Image Super-resolution:A Survey>,详细回顾了近年 ...
深度学习超分辨率最新综述：一文道尽技术分类与效果评测
置顶我爱计算机视觉,更快获取CVML新技术最近52CV介绍了好几篇图像超分辨率的工作,比如: CVPR 2019 神奇的超分辨率算法DPSR:应对图像模糊降质 CVPR 2019 | 旷视提出超分辨 ...
【深度学习】陶大程等人编写！最新41页深度学习理论综述
作者单位:悉尼大学(陶大程等人) 论文 https://arxiv.org/pdf/2012.10931.pdf 最近,悉尼大学陶大程等人发布了一个41页的最新深度学习理论的综述,对深度学习理论的最新 ...
100个深度图像分割算法，纽约大学UCLA等最新综述论文
转载自 https://new.qq.com/omn/20200122/20200122A0BEL300.html 100个深度图像分割算法,纽约大学UCLA等最新综述论文 [新智元导读]来自纽约大学 ...

关于深度学习理论和架构的最新综述(附录)