TensorFlow based Machine Learning VTU 2019 by pravin hanchinal
0 likes305 views
This document provides an overview of artificial intelligence and machine learning. It discusses what AI is, the differences between AI, machine learning, and deep learning. It also demonstrates some AI applications like self-driving cars, discusses the importance of AI, and reviews the types of machine learning. The document concludes by envisioning future applications of AI and discussing some challenges.
TensorFlow based Machine Learning VTU 2019 by pravin hanchinal
- 2. Disclaimer All the images are used for educational purposes only. All the images in the presentation are obtained from internet with “free to use license” search. Some of them are from free stock photos like pexels website. If you are owner of any images with copyright. Kindly report. Will remove the images with all respect to your creative work.
- 3. About Me Educator, Entrepreneur, Innovator, Cloud, Big Data, IoT, AI Evangelist pravinhanchinal.com
- 5. The Flow Session1: Introduction to AI and ML Session2: Demonstrations: Playing with AI Giorgio Cam Google Draw Session 3: Hands-on: Jupyter Notebook Google Colabs
- 6. What is AI? The science of intelligent programs
- 7. The Brain of AI | Neural Networks
- 8. Let’s See in Action: http://playground.tensorflow.org/
- 9. Why Now?
- 10. ‘Cos People are Stupid and Lazy
- 11. Smart Door, Still Stupid!
- 12. Why need Smart Cars...
- 17. AL vs ML vs DL Artificial Intelligence — Human Intelligence Exhibited by Machines Machine Learning — An Approach to Achieve Artificial Intelligence Deep Learning — A Technique for Implementing Machine Learning
- 18. What AI Deals with?
- 19. Why AI Important? 1. Reasoning: the ability to solve problems 2. Knowledge: the ability to represent knowledge about the world 3. Planning: the ability to set and achieve goals 4. Communication: the ability to understand written and spoken language. 5. Perception: the ability to deduce things from images, sounds, etc
- 20. What We Need for AI? Resources Tools Information
- 21. What We Have? Resources Enough Computing Power | Cheap Hardware Human Brains | Build AI Models Tools Cloud Open Source Tools Information Data
- 22. Agriculture Make AI learn based on our crop references, to predict pricing Classification of products using AI https://www.youtube.com/watch?v=MtQmqaSxszU
- 23. Working AI: Self Driving Cars https://youtu.be/tiwVMrTLUWg?t=7m56s
- 24. You Name it, We can apply it
- 25. Things Everywhere need ML
- 26. All our world is made of Things. Things when connected, form IoT IoT Won't Work Without Artificial Intelligence!
- 28. Machine Learning
- 30. How Machine Learning work?
- 31. Types of Machine Learning ❏ Supervised Learning – Train Me! ❏ Unsupervised Learning – I am self sufficient in learning ❏ Reinforcement Learning – My life My rules! (Hit & Trial)
- 38. In Future Drones Monitoring Farms and Harvesting Law Enforcement: Self Driving Cars Text Summarization in Law Automated RTI Toilet Doctors | Medical Insurance Office Less Services: Documents Movement Automated Citizen Care Prisoners monitoring
- 39. In Future Defence / Military Behavioural Analysis Predicting Poverty Guided Schooling
- 41. Let’s Play
- 43. Google Draw
- 45. Teachable Machine | Now https://teachablemachine.withgoogle.com/train/pose
- 49. Why ML for daily Life?
- 50. Walking
- 51. Running
- 52. Biking
- 53. Golf
- 56. Using Command line pip Use Docker image Using Jupyter Notebook Using Google Colab How to Use TensorFlow?
- 58. Google Colabs
- 59. Why Colabs? Google Colab comes with collaboration Colabs=Jupyter notebook+Google Docs Runs on Google servers (GPU) No Installation The notebooks are saved to Google Drive
- 61. Visioning AI Environment Birth get registered Aadhar, Insurance School suggestion->best friend College suggestion-> boy friend / girl friend Job suggestion Whom to marry When to have kids Old age home/Robot suggestion Instant call to Ambulance Finally your cemetery based on height
- 62. Sample Model : X-ROAD https://www.ria.ee/en/x-road.html
- 63. AI God God also knows before you pray
- 64. Best Research* in IoT/ML Uses Machine Learning Synthetic Sensors: http://www.gierad.com/projects/supersensor/ *read by me
- 65. The electric light did not come from the continuous improvement of candles. -Oren Harani
- 66. In a Nutshell Go through presentation again Explore resources provided Get Started Build Small Models and Scale up Learn From Failures
- 67. Finally to the tip of iceberg...
- 69. Thank you You gave me your time, the thoughtful gift of all...
- 73. Research References AI https://www.youtube.com/watch?v=8cmx7V4oIR8 https://www.youtube.com/watch?v=-Y7PLaxXUrs Self Driving Cars: https://www.youtube.com/watch?v=URmxzxYlmtg https://www.youtube.com/watch?v=OlLFK8oSNEM 10:38 http://sustain.stanford.edu/predicting-poverty/ https://interestingengineering.com/this-project-used-drones-to-farm-an-entire-crop-and-its-ready-for-harvesting
- 74. AI to Play Cam https://experiments.withgoogle.com/ai/giorgio-cam/view/ https://www.youtube.com/watch?v=eKeI63VSpto Draw https://quickdraw.withgoogle.com/ https://www.youtube.com/watch?time_continue=97&v=X8v1GWzZYJ4 AI Duet https://experiments.withgoogle.com/ai/ai-duet/view/ https://www.youtube.com/watch?time_continue=126&v=0ZE1bfPtvZo https://experiments.withgoogle.com/ai/drum-machine/view/
- 75. Challenges in AI Building Trust AI Human Interface Investment Software Malfunction Can Replace only Certain Tasks High Expectations
- 76. PEAS Representation Model P: Performance measure E: Environment A: Actuators S: Sensors