更改

大数据智能

删除3,842字节2018年7月29日 (日) 02:37
=人工智能简介=序言==
人工智能,是指计算机系统具备从听说读写到搜索、推理、决策和回答问题等类人智能的能力。即,感知、理解、决策的能力。技术科学的进步历程往往是理论通过实践开辟道路的过程。
==人工智能历史人工/机器智能 ==
过去经历了2次高潮与2次低谷[[人工智能]]/机器智能(Artificial / Machine Intelligence),是指计算机系统具备从听说读写到搜索、推理、决策和回答问题等类人智能的能力,即感知、理解、决策的能力。
网络和云计算所支持的计算能力[[人工智能实现思路]]
基于大数据的机器学习的算法进步=== 发展历史 ===
阅读材料:[[人工神经网络的历史]]
# The rebirth of neural networks-ISCA-2010 [http://pages.saclay.inria.fr/olivier.temam/homepage/ISCA2010web.pdf rebirth_NN]=实验竞赛数据集 =
==四个层面==[[实验数据集]]
===目标与功能分类=深度学习 ==
语音识别、机器视觉、自然语言理解。 深度学习(Deep Learning),机器学习中一种基于对数据进行表征学习的方法,试图使用包含复杂结构或由多重非线性变换构成的多个处理层对数据进行高层抽象的算法。
智能问答是综合以上功能的高级系统。[[深度学习]]
===核心技术分类=增强学习 ==
特定算法 机器学习算法 深度神经网络增强学习(Reinforcement Learning),是机器学习中的一个领域,强调如何基于环境而行动,以取得最大化的预期利益。
===软件工具===TensorFlow / Caffe / Torch[[增强学习]]
===底层硬件分类=机器感知 ==
可编程逻辑阵列 FPGA / 通用图形处理器 GPGPU / 通用处理器 CPU 群集机器感知(Machine Perception),如语音,图像,视频,手势,姿态等
==国际研究= 语音识别 ===
[http://research.google.com/teams/brain/ Googel_Brain]([http://research.google.com/pubs/jeff.html Jeff Dean语音识别]]),Automatic Speech Recognition,简称ASR
[https://research.facebook.com/ai Facebook_AI-Research](Yann LeCun)=== 计算机视觉 ===
[https://www.microsoft.com/en-us/research/group/dltc/ MSR Deep Learning Technology Center (DLTC)[计算机视觉](Li Deng)],Computer Vision,简称CV
[https://www.openai.com/blog/ OpenAI](Ilya Sutskever)== 机器认知 ==
==机器学习==机器认知(Machine Cognition),自然语言理解、推理、注意、知识、学习、决策、交互等。
Machine Learning'''技术手段:'''深度学习(Deep Learning)+ 增强学习(Reinforcement Learning)
[http://scikit-learn.org scikit-learn]=== 自然语言理解 ===
# Jordan, M. I., and T. M. Mitchell. "Machine learning: Trends, perspectives, and prospects." Science 349, no. 6245 (2015): 255-260. [http://science.sciencemag.org/content/349/6245/255 Machine_learning_science_2015]自然语言理解(Natural Language Understanding),使用的技术称为自然语言处理(Natural Language Processing,简称NLP)。
=深度学习== 智能问答 ===
Deep Neural Networks,简称DNN整合语音识别ASR,计算机视觉CV和自然语言处理NLP的问答系统QA。
[[卷积神经网络]]==语音合成==
[[深度学习-入门导读语音合成]]
==深度学习工具计算机游戏 ==
===谷歌===[[计算机游戏]]
[https://github.com/tensorflow/tensorflow TensorFlow]==机器翻译==
[http://download.tensorflow.org/paper/whitepaper2015.pdf Google_TensorFlow_whitepaper[机器翻译]]
===Facebook=推荐系统==
[https://github.com/facebook/fbcunn fbcunn][http://torch.ch/ Torch推荐系统]]
===微软===
 
[http://cntk.ai CNTK_Microsfot]
 
[https://www.microsoft.com/en-us/research/product/cognitive-toolkit/tutorials/ CNTK]
 
===DMLC===
 
[http://dmlc.ml/ dmlc]
 
[http://mxnet.io/ MXNet]
[https://github.com/dmlc/mxnet dmlc_mxnet]
 
===Université de Montréal===
 
 
[https://github.com/Theano/Theano/ Theano_code]
 
=增强学习=
增强学习(Reinforcement Learning)
 
[[增强学习-入门导读]]
 
==工具==
===谷歌===
deepmind/lab
[https://github.com/deepmind/lab deepmind_lab]
 
===OpenAI===
openai/universe
[https://github.com/openai/universe openai_universe]
 
=机器感知=
 
机器感知(Machine Perception),如语音,图像,视频,手势,姿态等
 
==<B>基于深度学习的机器感知</B>==
 
===语音识别===
 
语音识别(Automatic Speech Recognition),简称ASR
 
传统方法综述
 
:#Karpagavalli, S., and E. Chandra. "A Review on Automatic Speech Recognition Architecture and Approaches." International Journal of Signal Processing, Image Processing and Pattern Recognition 9, no. 4 (2016): 393-404.
 
基本工具
*:Long short term memory neural network (LSTM)
:#Long short term memory neural computation, Neural computation 9 (8), 1735-1780, 1997. [http://ieeexplore.ieee.org/document/6795963 LSTM]
 
*:Connectionist temporal classification (CTC)
:#Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks, ICML 2006.
 
*:Gated Recursive Unit (GRU)
:#On the Properties of Neural Machine Translation: Encoder-Decoder Approaches, SSST-8, 2014.
 
Alex Graves,DeepMind研究员,语音识别多项技术开创者。详见Google Scholar [https://scholar.google.com.hk/citations?user=DaFHynwAAAAJ Alex Graves]
:#Towards End-To-End Speech Recognition with Recurrent Neural Networks, ICML 2014.
:#Speech recognition with deep recurrent neural networks, 2013.
:#Hybrid speech recognition with deep bidirectional LSTM, ASRU 2013.
:#Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks, ICML 2006.
 
Google Speech
:# Google Speech Processing from Mobile to Farfield, CHiME 2016. [http://spandh.dcs.shef.ac.uk/chime_workshop/presentations/CHiME_2016_Bacchiani_keynote.pdf Google_Speech_Processing]
 
===计算机视觉===
 
计算机视觉(Computer Vision),简称 CV
 
* Object Detection
 
Ross Girshick, FAIR研究员,R-CNN的开创者。
 
[https://scholar.google.com.hk/citations?user=W8VIEZgAAAAJ Ross Girshick]
 
:<B>R-CNN (Region-based Convolutional Network method)</B>
::#Region based convolutional networks for accurate object detection and segmentation, TPAMI, 2015.
::#Rich feature hierarchies for accurate object detection and semantic segmentation, CVPR 2014.
 
:<B>Fast R-CNN (Fast Region-based Convolutional Network method)</B>
::#Fast R-CNN, ICCV 2015.
 
:<B>Faster R-CNN (Faster Region-based Convolutional Network method)</B>
::#Faster R-CNN Towards real-time object detection with region proposal networks, NIPS, 2015.
 
::• Fast_R-CNN(Python): https://github.com/rbgirshick/fast-rcnn
 
::• Faster_R-CNN(matlab): https://github.com/ShaoqingRen/faster_rcnn
 
::• Faster_R-CNN(Python): https://github.com/rbgirshick/py-faster-rcnn
 
=机器认知=
 
机器认知(Machine Cognition),自然语言理解,推理,注意,知识,学习,决策,交互等。
 
<B>技术手段:</B>
 
深度学习(Deep Learning) + 增强学习(Reinforcement/Unsupervised Learning)
 
=前沿应用进展=
 
==自然语言理解==
 
自然语言理解(Natural Language Understanding),使用的技术称为自然语言处理(Natural Language Processing,简称NLP)
 
==智能问答==
 
整合语音识别ASR,计算机视觉CV和自然语言处理NLP的问答系统QA。
Reasoning in vector space: An exploratory study of question answering, ICLR 2016.
=其它课程=相关资料==
相关课程:
[[实验室探究课-智能问答与智能系统]]
行政员管理员
6,105
个编辑