ai deep learning - cf machine learning

27
DEEP LEARNING Breakthrough In Machine Learning Intro & Tour

Upload: karl-seiler

Post on 17-Jan-2017

148 views

Category:

Software


0 download

TRANSCRIPT

Page 1: AI Deep Learning - CF Machine Learning

DEEP LEARNINGBreakthrough In Machine Learning

Intro & Tour

Page 2: AI Deep Learning - CF Machine Learning

Central Florida Machine Learning, Predictive Analytics and Automated Reasoning

“Smart tech is here, now, and integral to your engineering.”

Page 3: AI Deep Learning - CF Machine Learning

INTROS

Page 4: AI Deep Learning - CF Machine Learning

Exploring and sharing

How to…

What’s new / hot

Engineering Resources

Page 5: AI Deep Learning - CF Machine Learning

Data Science

Big Data

Machine Learning

Semantics

Predictive Analytics

Internet of Things

Artificial Intelligence

Language Processing

Page 6: AI Deep Learning - CF Machine Learning

DEEP LEARNINGBreakthrough In Machine Learning

Intro & Tour

Page 7: AI Deep Learning - CF Machine Learning

Bill Gates, then Chairman, Microsoft

“A breakthrough in machine learning would

be worth ten Microsofts.”

Page 8: AI Deep Learning - CF Machine Learning
Page 9: AI Deep Learning - CF Machine Learning
Page 10: AI Deep Learning - CF Machine Learning
Page 11: AI Deep Learning - CF Machine Learning

Deep Learning Algos

• Deep Boltzmann Machine (DBM)

• Deep Belief Networks (DBN)

• Convolutional Neural Network (CNN)

• Stacked Auto-Encoders

Page 12: AI Deep Learning - CF Machine Learning

HISTORY• Neural nets - big in the late 80’s -

Despite a commonly-held belief, there have been numerous successful applications

• Out of fashion in the 90’s

• 2003 renewed interest in the problem of learning representations (as opposed to just learning simple classifiers) - LeChun

• 2006-2007 traction via unsupervised training - Ng

• Now “Deep Learning” has come to designate any learning method that can train a system with more than 2 or 3 non-linear hidden layers.

Page 13: AI Deep Learning - CF Machine Learning

WHY NOW?• More and diverse data

• More processing power

• Algorithm advances and discoveries

• GPUs

Page 14: AI Deep Learning - CF Machine Learning

Andrew Ng Yann

LeCun

Page 15: AI Deep Learning - CF Machine Learning

EXAMPLES

Page 16: AI Deep Learning - CF Machine Learning

Image Recognition

Page 17: AI Deep Learning - CF Machine Learning

Image Recognition

Page 18: AI Deep Learning - CF Machine Learning

Speech Recognition

Page 19: AI Deep Learning - CF Machine Learning

Natural Language

• Constituency parsing

• Sentiment analysis

• Information retrieval

• Machine translation

• Contextual entity linking

Page 20: AI Deep Learning - CF Machine Learning

VIDEO

Page 21: AI Deep Learning - CF Machine Learning

Andrew Ng: Deep Learning…

https://www.youtube.com/watch?v=n1ViNeWhC24

Page 22: AI Deep Learning - CF Machine Learning

Introduction to Deep Learning with Python

https://www.youtube.com/watch?v=S75EdAcXHKk

Page 23: AI Deep Learning - CF Machine Learning

TOOLS

Page 24: AI Deep Learning - CF Machine Learning

TensorFlow™ is an open source software library for numerical computation using data flow graphs. Nodes in the graph represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) communicated between them. The flexible architecture allows you to deploy computation to one or more CPUs or GPUs in a desktop, server, or mobile device with a single API.

Page 25: AI Deep Learning - CF Machine Learning

The goal of Torch is to have maximum flexibility and speed in building your scientific algorithms while making the process extremely simple. Torch comes with a large ecosystem of community-driven packages in machine learning, computer vision, signal processing, parallel processing, image, video, audio and networking among others, and builds on top of the Lua community.

At the heart of Torch are the popular neural network and optimization libraries which are simple to use, while having maximum flexibility in implementing complex neural network topologies. You can build arbitrary graphs of neural networks, and parallelize them over CPUs and GPUs in an efficient manner.

Page 26: AI Deep Learning - CF Machine Learning

• Pylearn2 is a machine learning library. Most of its functionality is built on top of Theano. It provides parallelization with CPUs and GPUs.

• Theano — An open source machine learning library for Python.• Deeplearning4j — An open source deep learning library written for Java. It provides parallelization with

CPUs and GPUs.• OpenNN — An open source C++ library which implements deep neural networks and provides

parallelization with CPUs.• NVIDIA cuDNN — A GPU-accelerated library of primitives for deep neural networks.• DeepLearnToolbox — A Matlab/Octave toolbox for deep learning.• convnetjs — A Javascript library for training deep learning models. It contains online demos.• Gensim — A toolkit for natural language processing implemented in the Python programming language.• Caffe — A deep learning framework.• Apache SINGA — A deep learning platform developed for scalability, usability and extensibility.• RNNLM — RNN language model open source.• RNNLMPara — Parallel RNN language model trainer open source.

Other Tools

Page 27: AI Deep Learning - CF Machine Learning

Karl Seiler | [email protected]@pivitguruSMARTER CHANGE