20
Deep Learning Tutorial (I) Introduction Guan Wang

Deep learning tutorial (i)

Embed Size (px)

Citation preview

Page 1: Deep learning tutorial (i)

Deep Learning Tutorial (I)IntroductionGuan Wang

Page 2: Deep learning tutorial (i)

Contents

Overview

Industry Landscape

System architectures

Bleeding edges

Page 3: Deep learning tutorial (i)

Overview

Page 4: Deep learning tutorial (i)

Deep neural networks learn to do the following

Page 5: Deep learning tutorial (i)

Yearbook• prior 2012

• Machine learning was more about SVM, Graphical Models, non-parametric baysian, and simple hacks, e.g., decision tree, naïve bayesian, regressions, etc

•2012~2013

• Deep Neural Network, big data, mature distributed computing architectures

• Refreshing accuracy record in image recognition tasks

• Feed forward Neural Networks, CNN, RBM

•2014~2015

• Bridging CV, NLP and expanding to other domains

• New architectures and new ML tasks

• RNN, LSTM, RL

•2016~future

• Larger models on CV, NLP, keep expanding to other domains

• Larger systems, larger models on CNN, LSTM, etc

Page 6: Deep learning tutorial (i)

Industry Landscape

Page 7: Deep learning tutorial (i)

Hardware Optimized for Neural Networks

Google TPUNVIDIA

Page 8: Deep learning tutorial (i)

Deep Learning Service on cloud

Page 9: Deep learning tutorial (i)

Computer Vision (crowded market)Generic Algorithms & API providers•Special object recognition (face, etc)•General object recognition (image search, etc)•Moving object detection & recognition (pedestrian detection, etc)•Image understanding (visual QA, artify, etc)•Video understanding (video search, etc)Verticals•Satellite image analysis (understanding civil developments)•Home & office place security & surveillance•User interest analysis and high precision targeting (ads)•ADAS & Autonomous Driving•Robotics and drones•The list goes on and on

Page 10: Deep learning tutorial (i)

Natural Language Understanding (crowded market)

Generic Algorithms & API providers•Personal assistance (x.ai, api.ai, etc)•Chat bots (facebook ecosystem, viv.ai, etc)•Knowledge understanding (IBM watson, etc)

Verticals•Customer service•Travel management•Financial service•Smart homes•Connected cars•The list goes on and on

Page 11: Deep learning tutorial (i)

System Architecture

Every good machine learning algorithm deserves its own system architecture. -- one of my mentors

Page 12: Deep learning tutorial (i)

Distributed ArchitecturesGeneric solvers•Stochastic Gradient Descent•Coordinate Descent•MCMC•ADMM•…Design Choices•sync or async•CPU or GPU cluster•Online training or offline training•...Examples•Parameter Server•DistBelief•Tensorflow

Flexible solutions•CoreOS, etcd, Docker•Kubernetes•Pachyderm•Mesos•etcBig data ecosystem•Spark & tachyon•yarn•hdfs•etcDeep learning tools•Tensorflow•Torch-IPC•DistBelief•Petuum•GraphLab

Page 13: Deep learning tutorial (i)

Standalone ToolkitsCPU with GPU speedup •Torch•Caffe•Theano•Tensorflow

Good for research or small applications

Embedded Support•Tiny-CNN•Tensorflow-embedded

Good for phone apps, raspberry Pi, cars, drones, robotics

Page 14: Deep learning tutorial (i)

Bleeding edge directions

Page 15: Deep learning tutorial (i)

CNN

CNN on Text ClassificationR-CNN, fast R-CNNAttention Model

Page 16: Deep learning tutorial (i)

LSTMSeq2seq model•Translation•Dialogue system•Time series analysis•Text classification

Page 17: Deep learning tutorial (i)

Memory NetworksExternal memory on RNN•Translation•Dialogue system•QA

Page 18: Deep learning tutorial (i)

Reinforcement LearningDeep Q-LearningFunction approximation on•Policy function•Value function

Page 19: Deep learning tutorial (i)

Deep Generative ModelAdversarial networks

Page 20: Deep learning tutorial (i)

The End

Thank You!