This document appears to be a presentation about artificial intelligence and deep learning concepts. It discusses topics like neural networks, gradient descent, and how deep learning uses neural networks to automatically learn representations of raw data through backpropagation with just one line of code. It provides examples of neural network concepts like loss functions, weights, and activation functions. The presentation aims to explain these complex AI concepts in a simple and intuitive way for all levels of understanding. It includes acknowledgments and credits to other resources that inspired or informed the content.
Semantic Optimization with Structured Data - SMX MunichCraig Bradford
Craig Bradford, SMX Munich
The future of structured data isn’t about understanding what a thing is, it’s about understanding what a thing can do. It is allowing us to move past "strings to things" and into actions and anticipatory search. In this presentation I cover:
How structured data has changed (strings to things)
How to get apps indexed by Google
Using structured data to say what a thing can do (things to actions)
Email markup for events and more
The future of Google Now (actions to anticipation)
Some predictions and trends about what comes next
Google announced its new Hummingbird algo update, putting lots of SEOs into doubts and concerns – what should this algo change mean to you and your website? To make sure you don't get down into panic and plan your SEO strategy wisely, we've put up this short guide to explain what Hummingbird update is, how it affects your rankings and how to adapt your SEO strategy to benefit from the changes.
Everyone wants their website to rank at #1 in a Google search but after writing their site’s content they don’t know the next steps for competing with the other 1 billion+ websites on the world wide web.
Did you know that Google has a Keyword Planner tool that tells you how much or how little competition a certain search phrase will yield? Do you use a standard naming convention for files and media that you upload or embed? Are you ensuring that your CMS is generating the right HTML tags for your content? There are several simple steps you may be missing when it comes to optimizing your website for search engines.
Instead of immediately shelling out hundreds of dollars for an SEO strategist, take a deep breath and then implement these often under-utilized tricks for improving your organic search engine ranking. With time, you’ll find the traffic you’re looking for.
Takeaways: I want attendees to take home 5-10 new ideas for boosting their website’s search engine ranking. I’m hoping to keep these ideas accessible to non-tech/non-HTML workers but hopefully I can inspire some web programmers to rethink some of their own processes when working on their company’s or clients’ websites.
Attendee skill level: either some experience with content strategy writing or project management; a newcomer to web development or design
---
Presented at MinneWebCon May 1, 2017 in Minneapolis, MN (http://minnewebcon.org/)
As search evolves, so does optimization. Search results are less about phrases (combinations of words and letters) and more about topics (semantic meanings and entities). So a smart content marketer optimizes for “things, not strings.”
But what exactly does this mean for the writer? In this presentation Andy Crestodina will cover five specific actions we take as content marketers to make sure that your marketing is aligned with the future of SEO.
Find clues into what topics are semantically linked to each other (Research)
Target topics, not just phrases, through writing (Semantic Search)
Incorporate natural language into your content (Voice Search)
Make visitors happy in ways that make Google happy (User Interaction Signals)
You’re about to learn the step-by-step process for each of the specific actions that will future-proof your search engine rankings.
Semantic Optimization with Structured Data - SMX MunichCraig Bradford
Craig Bradford, SMX Munich
The future of structured data isn’t about understanding what a thing is, it’s about understanding what a thing can do. It is allowing us to move past "strings to things" and into actions and anticipatory search. In this presentation I cover:
How structured data has changed (strings to things)
How to get apps indexed by Google
Using structured data to say what a thing can do (things to actions)
Email markup for events and more
The future of Google Now (actions to anticipation)
Some predictions and trends about what comes next
Google announced its new Hummingbird algo update, putting lots of SEOs into doubts and concerns – what should this algo change mean to you and your website? To make sure you don't get down into panic and plan your SEO strategy wisely, we've put up this short guide to explain what Hummingbird update is, how it affects your rankings and how to adapt your SEO strategy to benefit from the changes.
Everyone wants their website to rank at #1 in a Google search but after writing their site’s content they don’t know the next steps for competing with the other 1 billion+ websites on the world wide web.
Did you know that Google has a Keyword Planner tool that tells you how much or how little competition a certain search phrase will yield? Do you use a standard naming convention for files and media that you upload or embed? Are you ensuring that your CMS is generating the right HTML tags for your content? There are several simple steps you may be missing when it comes to optimizing your website for search engines.
Instead of immediately shelling out hundreds of dollars for an SEO strategist, take a deep breath and then implement these often under-utilized tricks for improving your organic search engine ranking. With time, you’ll find the traffic you’re looking for.
Takeaways: I want attendees to take home 5-10 new ideas for boosting their website’s search engine ranking. I’m hoping to keep these ideas accessible to non-tech/non-HTML workers but hopefully I can inspire some web programmers to rethink some of their own processes when working on their company’s or clients’ websites.
Attendee skill level: either some experience with content strategy writing or project management; a newcomer to web development or design
---
Presented at MinneWebCon May 1, 2017 in Minneapolis, MN (http://minnewebcon.org/)
As search evolves, so does optimization. Search results are less about phrases (combinations of words and letters) and more about topics (semantic meanings and entities). So a smart content marketer optimizes for “things, not strings.”
But what exactly does this mean for the writer? In this presentation Andy Crestodina will cover five specific actions we take as content marketers to make sure that your marketing is aligned with the future of SEO.
Find clues into what topics are semantically linked to each other (Research)
Target topics, not just phrases, through writing (Semantic Search)
Incorporate natural language into your content (Voice Search)
Make visitors happy in ways that make Google happy (User Interaction Signals)
You’re about to learn the step-by-step process for each of the specific actions that will future-proof your search engine rankings.
DevLearn 2018 - Designing AR Experiences for Performance SupportChad Udell
While many companies are experimenting with AR in the L&D space, there are a number of businesses harnessing the power of AR for enhancing operational performance outside of the training department. How do these experiences differ, and how can you renew your department’s focus on performance by taking on more advanced AR solutions in your efforts?
In this session, you will learn practical approaches for designing effective AR experiences. You’ll discover an approach to strategic implementation of AR by forming a partnership with functional business units. You’ll also explore the difference between simple marker-based AR solutions and more advanced computer vision and machine learning–backed AR. You’ll then look at how you can integrate AR systems with operational business systems in order to maximize return on investment and realize the opportunity that AR-enabled workers represent. Finally, you’ll look at aligning measurement of business task success and AR experience usage in order to align learning and production.
RSC SE Teaching toolkit no 8 Todaysmeet, QR codes and Slideshare - Jane Macke...Jane Mackenzie
The 8th Teaching Toolkit Workshop as part of the Online Innovation Week. Discover how to use these 3 free tools. There will be a recording of the session available on the RSC SE website in 2 weeks. http://www.jiscrsc.ac.uk/southeast
Getting SEO done - Why thinking agile is the best SEO skill in 2020 - SEOday ...Charlie Williams
In this talk at SEOday in January 2020, I spoke to the audience about how the best SEO (& content strategy) skill isn't a coding language, keyword research technique or digital PR tactic. It's the ability to get your SEO recommendations actually put in place. To have a mindset of constant improvement & working towards a 'perfect' website - something that's not achievable, but gets you in the right mindset.
#CRODay - User Identification and Psychology: the New Way to Convert in 2015Natzir Turrado
Personalisation. User Activation. Recommendation. Multi device tracking. What do they all have in common? Users need to Identify themselves.
When our users register and login in our websites, we can increase conversions by using powerful techniques based on our users preference, experience or behaviour.
Converting anonymous sessions into registered users (user Registration & Login) has never been a priority of web teams. It's time to change this and increase conversions in a brand new way. The 2015 way.
In this talk Xavier and Natzir, two skilled and veteran CRO's, will discuss this topic and give you tips to increase the login rate on your website. This will allow you to increase your conversion rates by personalising your user's experience.
Presentation about SEO for IAB Belgium @Google Offices BXL (intermediate level)
TOC:
- The SEO Pyramid
- Which ranking factors matter
- SEO trends
- SEO Migration
- New sites & SEO
- Social Media & SEO
- International SEO
- Local SEO
- Video & Image SEO
- Keyword Research (finally done right)
- Optimizing your website / writing content
How to Create Fun User Experience by Shutterstock Dir of ProductProduct School
Main takeaways:
- Understand the value of your product, it’s corresponding emotions, and deliver it as quickly as possible
- Break down each part of your customer journey in order to inject and heighten the value you’re providing
- Utilize qualitative data to discover opportunities for joy that don’t distract from your main value
Using the Google Analytics API to make most popular pages widgets for your we...Dean Peters
Using the Google Analytics API to create a most popular pages widget for your website or blog using the Google Analytics API. I'll demonstrate a custom report in Google Analytics, then show how to automate the same using Perl and the Google Analytics API to create an RSS feed of the top 10 stories from my blog — which I then incorporate into the siderail of my blog using a WordPress RSS widget.
Here's the link to code in GitHub:
http://bitly.com/ga-api2mpp?slideshare
Honest Reviews of Tim Han LMA Course Program.pptxtimhan337
Personal development courses are widely available today, with each one promising life-changing outcomes. Tim Han’s Life Mastery Achievers (LMA) Course has drawn a lot of interest. In addition to offering my frank assessment of Success Insider’s LMA Course, this piece examines the course’s effects via a variety of Tim Han LMA course reviews and Success Insider comments.
DevLearn 2018 - Designing AR Experiences for Performance SupportChad Udell
While many companies are experimenting with AR in the L&D space, there are a number of businesses harnessing the power of AR for enhancing operational performance outside of the training department. How do these experiences differ, and how can you renew your department’s focus on performance by taking on more advanced AR solutions in your efforts?
In this session, you will learn practical approaches for designing effective AR experiences. You’ll discover an approach to strategic implementation of AR by forming a partnership with functional business units. You’ll also explore the difference between simple marker-based AR solutions and more advanced computer vision and machine learning–backed AR. You’ll then look at how you can integrate AR systems with operational business systems in order to maximize return on investment and realize the opportunity that AR-enabled workers represent. Finally, you’ll look at aligning measurement of business task success and AR experience usage in order to align learning and production.
RSC SE Teaching toolkit no 8 Todaysmeet, QR codes and Slideshare - Jane Macke...Jane Mackenzie
The 8th Teaching Toolkit Workshop as part of the Online Innovation Week. Discover how to use these 3 free tools. There will be a recording of the session available on the RSC SE website in 2 weeks. http://www.jiscrsc.ac.uk/southeast
Getting SEO done - Why thinking agile is the best SEO skill in 2020 - SEOday ...Charlie Williams
In this talk at SEOday in January 2020, I spoke to the audience about how the best SEO (& content strategy) skill isn't a coding language, keyword research technique or digital PR tactic. It's the ability to get your SEO recommendations actually put in place. To have a mindset of constant improvement & working towards a 'perfect' website - something that's not achievable, but gets you in the right mindset.
#CRODay - User Identification and Psychology: the New Way to Convert in 2015Natzir Turrado
Personalisation. User Activation. Recommendation. Multi device tracking. What do they all have in common? Users need to Identify themselves.
When our users register and login in our websites, we can increase conversions by using powerful techniques based on our users preference, experience or behaviour.
Converting anonymous sessions into registered users (user Registration & Login) has never been a priority of web teams. It's time to change this and increase conversions in a brand new way. The 2015 way.
In this talk Xavier and Natzir, two skilled and veteran CRO's, will discuss this topic and give you tips to increase the login rate on your website. This will allow you to increase your conversion rates by personalising your user's experience.
Presentation about SEO for IAB Belgium @Google Offices BXL (intermediate level)
TOC:
- The SEO Pyramid
- Which ranking factors matter
- SEO trends
- SEO Migration
- New sites & SEO
- Social Media & SEO
- International SEO
- Local SEO
- Video & Image SEO
- Keyword Research (finally done right)
- Optimizing your website / writing content
How to Create Fun User Experience by Shutterstock Dir of ProductProduct School
Main takeaways:
- Understand the value of your product, it’s corresponding emotions, and deliver it as quickly as possible
- Break down each part of your customer journey in order to inject and heighten the value you’re providing
- Utilize qualitative data to discover opportunities for joy that don’t distract from your main value
Using the Google Analytics API to make most popular pages widgets for your we...Dean Peters
Using the Google Analytics API to create a most popular pages widget for your website or blog using the Google Analytics API. I'll demonstrate a custom report in Google Analytics, then show how to automate the same using Perl and the Google Analytics API to create an RSS feed of the top 10 stories from my blog — which I then incorporate into the siderail of my blog using a WordPress RSS widget.
Here's the link to code in GitHub:
http://bitly.com/ga-api2mpp?slideshare
Honest Reviews of Tim Han LMA Course Program.pptxtimhan337
Personal development courses are widely available today, with each one promising life-changing outcomes. Tim Han’s Life Mastery Achievers (LMA) Course has drawn a lot of interest. In addition to offering my frank assessment of Success Insider’s LMA Course, this piece examines the course’s effects via a variety of Tim Han LMA course reviews and Success Insider comments.
Francesca Gottschalk - How can education support child empowerment.pptxEduSkills OECD
Francesca Gottschalk from the OECD’s Centre for Educational Research and Innovation presents at the Ask an Expert Webinar: How can education support child empowerment?
Unit 8 - Information and Communication Technology (Paper I).pdfThiyagu K
This slides describes the basic concepts of ICT, basics of Email, Emerging Technology and Digital Initiatives in Education. This presentations aligns with the UGC Paper I syllabus.
Introduction to AI for Nonprofits with Tapp NetworkTechSoup
Dive into the world of AI! Experts Jon Hill and Tareq Monaur will guide you through AI's role in enhancing nonprofit websites and basic marketing strategies, making it easy to understand and apply.
How to Make a Field invisible in Odoo 17Celine George
It is possible to hide or invisible some fields in odoo. Commonly using “invisible” attribute in the field definition to invisible the fields. This slide will show how to make a field invisible in odoo 17.
Macroeconomics- Movie Location
This will be used as part of your Personal Professional Portfolio once graded.
Objective:
Prepare a presentation or a paper using research, basic comparative analysis, data organization and application of economic information. You will make an informed assessment of an economic climate outside of the United States to accomplish an entertainment industry objective.
The Roman Empire A Historical Colossus.pdfkaushalkr1407
The Roman Empire, a vast and enduring power, stands as one of history's most remarkable civilizations, leaving an indelible imprint on the world. It emerged from the Roman Republic, transitioning into an imperial powerhouse under the leadership of Augustus Caesar in 27 BCE. This transformation marked the beginning of an era defined by unprecedented territorial expansion, architectural marvels, and profound cultural influence.
The empire's roots lie in the city of Rome, founded, according to legend, by Romulus in 753 BCE. Over centuries, Rome evolved from a small settlement to a formidable republic, characterized by a complex political system with elected officials and checks on power. However, internal strife, class conflicts, and military ambitions paved the way for the end of the Republic. Julius Caesar’s dictatorship and subsequent assassination in 44 BCE created a power vacuum, leading to a civil war. Octavian, later Augustus, emerged victorious, heralding the Roman Empire’s birth.
Under Augustus, the empire experienced the Pax Romana, a 200-year period of relative peace and stability. Augustus reformed the military, established efficient administrative systems, and initiated grand construction projects. The empire's borders expanded, encompassing territories from Britain to Egypt and from Spain to the Euphrates. Roman legions, renowned for their discipline and engineering prowess, secured and maintained these vast territories, building roads, fortifications, and cities that facilitated control and integration.
The Roman Empire’s society was hierarchical, with a rigid class system. At the top were the patricians, wealthy elites who held significant political power. Below them were the plebeians, free citizens with limited political influence, and the vast numbers of slaves who formed the backbone of the economy. The family unit was central, governed by the paterfamilias, the male head who held absolute authority.
Culturally, the Romans were eclectic, absorbing and adapting elements from the civilizations they encountered, particularly the Greeks. Roman art, literature, and philosophy reflected this synthesis, creating a rich cultural tapestry. Latin, the Roman language, became the lingua franca of the Western world, influencing numerous modern languages.
Roman architecture and engineering achievements were monumental. They perfected the arch, vault, and dome, constructing enduring structures like the Colosseum, Pantheon, and aqueducts. These engineering marvels not only showcased Roman ingenuity but also served practical purposes, from public entertainment to water supply.
Palestine last event orientationfvgnh .pptxRaedMohamed3
An EFL lesson about the current events in Palestine. It is intended to be for intermediate students who wish to increase their listening skills through a short lesson in power point.
Instructions for Submissions thorugh G- Classroom.pptxJheel Barad
This presentation provides a briefing on how to upload submissions and documents in Google Classroom. It was prepared as part of an orientation for new Sainik School in-service teacher trainees. As a training officer, my goal is to ensure that you are comfortable and proficient with this essential tool for managing assignments and fostering student engagement.
2. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Topics in this presentation
Basic conceptual understanding of
Neural Networks, Gradient Descent
3. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
New way to think from today !
4. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
5. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Represent the same data in a new way
•
6. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Represent the same data in a new way
•
7. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Represent the same data in a new way
•
8. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Find a way to represent the data in a
better way?
Deep learning is === a multistage way to learn data representations.
It’s a simple idea—but, as it turns out, very simple mechanisms, sufficiently scaled, can end up looking like magic.
All this happens with just 1 line of code!
9. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Just 1 line of code == magic!
Deep learning is === a multistage way to learn data representations.
It’s a simple idea—but, as it turns out, very simple mechanisms, sufficiently scaled, can end up looking like magic.
myModel. fit ( X , Y )
10. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
All this happens with just 1 line of code!
11. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
How did the magic happen?
)
)
)
)
(x, y) = load_data()
12. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Can a neural network automatically represent
the raw data into another way?
13. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Deep representations learned by a
neural network model
14. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
representations “learned” automatically
15. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
A neural network is parameterized by its
weights
16. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
A loss function measures the quality of the
network’s output
The loss function takes the predictions
of the network and the true target.
and computes a distance score,
capturing how well the network has
done on this specific example
Credits, Deep Learning with Python
17. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
The loss score is used as a feedback signal to
adjust the weights
18. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
•
Credits: Luis Serrano
https://www.youtube.com/watch?v=BR9h47Jtqyw
19. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
20. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Teach a computer how to split the data
21. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
22. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
23. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
24. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Group Activity game to learn
Gradient Descent
•
25. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
26. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
4 Graident descent 1145 game
Credits: Luis Serrano
https://www.youtube.com/watch?v=BR9h47Jtqyw
27. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
28. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Gradient descent
29. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
30. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Credits: Google Machine Learning Crash Course
https://developers.google.com/machine-learning/crash-course/
31. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
32. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
33. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Credits: Google Machine Learning Crash Course
https://developers.google.com/machine-learning/crash-course/
34. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
35. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
36. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
37. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
38. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
39. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
What should be Learning rate ?
• https://developers.google.com/machine-learning/crash-course/fitter/graph
40. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Neural network is inspired by biology,
In reality, it just a mathematical function
41. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Neuron === Perceptron
•
42. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Switches On/Off a nerve connection
43. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Switches On/Off a nerve connection
44. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Simply an weighted average
•
Sum = 3.3
45. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Then a “activation function”
•
46. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
•
47. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
•
48. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
•
49. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Multi-Layer Perceptron (MLP)
•
50. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Deep Neural Network
•
51. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Design the network architecture
52. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Design the network architecture
network.add( keras.layers.Dense(512))
network.add( keras.layers.Dense(128))
network.add( keras.layers.Dense(10))
network.add( keras.layers.Dense(512))
network.add( keras.layers.Dense(10))
53. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
54. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
55. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
56. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
57. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
58. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
59. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
60. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Design of Deep Learning experiment
Define problem What is the purpose
Prepare the data
Collect data, spilit data,
vectorize, reshape,
normalize, OHE
Define the network
architecture
architecture
Define loss,
optimizers
Define the metrics,
optimizers, loss function,
and Compile the model
Train the network Train the network
Improve power to
generalize
Evaluate the network,
Hypertune
Test Test, Predict
Deploy Deploy
1
2
3
4
5
6
7
8AIforEveryone.org
61. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Method: 7 steps to develop your Deep Learning model
1: Define problem
Collect data representing
the purpose
2: Format the data
Spilt data, vectorize, reshape,
normalize, OHE
3: Design the network Design the Neural Network
4: Define loss Think of what to optimize for
5: Train the network
Allow the network to learn
patterns in the data
6: Validate & Improve Power of generalization?
7: Predict Predict
Adjust model’s
capacity to learn
“just the patterns”
Develop model
that overfits
Develop 1st model
62. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
(trainX, trainY), (testX, testY) = mnist.load_data()
2. Prepare
the data
Spilt data in
training &
testing
Vectortize
as tensors
Reshape Normalize
One Hot
encode
No of training records = 60000
No of test records = 10000
63. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Multi dimensional Array == Tensor
What_is_size_of_array = myArray.shape
64. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
• 1D Tensor
• Array
• 2D Tensor
• 2D array
• Matrix
• 3D Tensor
• 3D array
• 4D Tensor
•
• 5D Tensor
65. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
One Hot Encoding
Data:
[2]
Data after One Hot Encoding:
[[ 0. 0. 1. 0. 0. 0. 0. 0. 0. 0.]]
yNew = to_categorical (y)
66. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
2. Prepare
the data
Spilt data in
training &
testing
Vectortize
as tensors
Reshape Normalize
One Hot
encode2. Prepare
the data
Spilt data in
training &
testing
Vectortize
as tensors
Reshape Normalize
One Hot
encode
myArray = keras.utils. to_categorical ( aInteger)
Data: [2]
Data after OHE : [[ 0. 0. 1. 0. 0. 0. 0. 0. 0. 0.]]
67. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
7 easy steps to develop your 1st Deep Learning model
1: Define
problem
Load the collected data,
Visualize it
2: Prepare
the data
Spilt data, vectorize,
reshape, normalize, OHE
3: Define the
network
architecture
Design the Network
4: Define
optimizers
Define the metrics,
optimizers, loss function,
and Compile the model
5: Train the
network
Train the network
6: Validate &
Improve
model
Evaluate the network
7: Predict Predict
Adjust model’s capacity
to learn
“just the patterns”
Develop model that
overfits
Develop 1st model
AIforEveryone.org
68. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
69. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
3. Design the network
AIforEveryone.org
Adjust model’s capacity
to learn
“just the patterns”
Develop model that
overfits
Develop 1st model
Data
RepresentationofData
70. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
3. Design the network
Adjust model’s capacity
to learn
“just the patterns”
Develop model that
overfits
Develop 1st model
Limit the capacity
so that only the
representations can
be learnt
71. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
5. Train
4 4 4
72. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
73. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
7. Predict
Hand
drawing
samples
0
1
.
9
Classifier
Input Predicted output
74. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
7. Predict
Problem: Users draws a hand drawn gestures , your app predicts it.
predictedResults = network.predict(testXready[1:3])
Predicted Results:
[ 1.35398892e-08 3.91871922e-08 9.99876499e-01 1.22501457e-04
1.77109036e-10 1.30937892e-07 2.99046292e-07 2.58880729e-11
4.58700157e-07 2.09212196e-11]
Ground Truth OHE : [[ 0. 0. 1. 0. 0. 0. 0. 0. 0. 0.]]
Ground Truth : [2]
AIforEveryone.org
75. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Let’s it together
• App idea: Recognize your hand gesture to
unlock your smartphone (Secure unlock)
• Biz outcome: Recognize hand draw gestures
to unlock your phone
Test accuracy is: 0.9157
#Step 6: Evaluate the network
metrics_test_loss, metrics_test_accuracy = network.evaluate(testXready, testYready)
76. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Design of Deep Learning experiment
Define problem What is the purpose
Prepare the data
Collect data, spilit data,
vectorize, reshape,
normalize, OHE
Define the network
architecture
architecture
Define loss,
optimizers
Define the metrics,
optimizers, loss function,
and Compile the model
Train the network Train the network
Improve power to
generalize
Evaluate the network,
Hypertune
Test Test, Predict
Deploy Deploy
1
2
3
4
5
6
7
8
77. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Experiment 2
• MNIST_Predict_with_Dense_and_Tune_Capacity
78. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
•
79. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
80. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Our inspiration
Friendly approaches :
1) KERAS.io
François Chollet’s
Book on “Deep Learning with Python”
2) Deeplearning.ai (Coursera.org)
Andrew Ng
3) Google Machine Learning Course
https://developers.google.com/machine-learning/crash-course/
81. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
More inspiration
Excellent Resources
• Stanford cs231 n
http://cs231n.stanford.edu
• MIT Deep Learning
http://introtodeeplearning.com/
https://deeplearning.mit.edu
• IIT Madras
my classes notes with Prof. Anurag (Deep Learning)
82. Acknowledgments & Credits are mentioned to inspirational resources presented in the end of presentation. For more like this, https://sites.google.com/view/AIforEveryone
Want a Book?
Francois Chollet
Artificial Intelligence Researcher,
Google
ISBN-13: 978-1617294433
Deep Learning with Python
https://www.manning.com/books/deep-learning-with-python
Book Published: December 22, 2017
Check free chapters at Manning Website!