AI - To get an overview
- 1. School of AI
- 2. AI is the New Electricity AI 2
- 3. Artificial Intelligence • What is AI • Artificial Intelligence is a way to make machines think and behave intelligently. • Machines understand the World and accordingly react to situations in the same way that Humans do • AI is closely related to the study of Human Brain. AI 3
- 4. Why do we need to study AI? • AI has the ability to impact every aspect of our lives. • The field of AI tries to understand patterns and behaviors of entities. • With AI, we want to build smart systems and understand the concept of intelligence as well. • The intelligent systems that we construct are very useful in understanding how an intelligent system like our brain goes about constructing another intelligent system. AI 4
- 5. How Brain Thinks AI 5
- 6. How AI Thinks AI 6
- 7. Applications of AI • Computer Vision • Natural Language Processing • Speech Recognition • Expert Systems • Games • Robotics AI 7
- 9. Natural Language Processing • Spell checking, keyword search, finding synonyms • Extracting information from websites such as • product price, dates, location, people or company names checking, keyword search, finding synonyms • Machine translation • Spoken dialog systems • Complex question answering • Automated/assisted translation • Sentiment analysis for marketing or finance/trading • Speech recognition • Chatbots / Dialog agents AI 9
- 10. AI / ML / DL AI 10
- 11. AI / ML / DL AI 11
- 12. Machine Learning • Using Data to Answer Questions Training Prediction Using Answer Data Questions AI 12
- 13. Machine Learning Process Flow AI 13
- 15. Machine Learning – Supervised Learning • Supervised Learning • Given a set of data points {x(1),...,x(m)} associated to a set of outcomes {y(1),...,y(m)}, we want to build a classifier that learns how to predict y from x. • Type of prediction ― The different types of predictive models are summed up in the table below: AI 15
- 16. Machine Learning – Supervised Learning • Type of Model ― The different models are summed up in the table below: AI 16
- 17. Support Vector Machines • The goal of support vector machines is to find the line that maximizes the minimum distance to the line • Optimal margin classifier – The optimal margin classifier ’h’ in such that: AI 17
- 18. Naïve Bayes • Assumption – The Naïve Bayes model supposes that the feature of each data point are all independent. • AI 18
- 19. Machine Learning – Unsupervised Learning • The goal of unsupervised learning is to find hidden patterns in unlabeled data {x(1),...,x(m)}. • Jensen's inequality ― Let g be a convex function and X a random variable. We have the following inequality: AI 19
- 20. Unsupervised Learning – k-means Clustering • The AI 20
- 21. Unsupervised Learning – k-means Clustering • The AI 21
- 24. Supervised Learning Vs Unsupervised Learning AI 24
- 25. ML Algorithm – Linear Regression AI 25
- 27. Machine Learning Model Selection AI 27
- 28. ML Algorithm – Logistic Regression AI 28
- 29. ML Algorithm – Classification and Regression Trees AI 29
- 30. ML Algorithm – Naïve Bayes AI 30
- 31. ML Algorithm – K – Nearest Neighbours AI 31
- 32. ML Algorithm – Support Vector Machines AI 32
- 33. Deep Learning • Neural Networks and Deep Learning • Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization • Structuring Machine Learning Projects • Convolutional Neural Networks • Sequence Models AI 33
- 36. Deep Learning - CNN AI 36
- 37. Deep Learning – Neural Network AI 37
- 38. TensorFlow • TensorFlow™ is an open source software library for high performance numerical computation. • Its flexible architecture allows easy deployment of computation across a variety of platforms (CPUs, GPUs, TPUs), and from desktops to clusters of servers to mobile and edge devices. • Originally developed by researchers and engineers from the Google Brain team within Google’s AI organization, it comes with strong support for machine learning and deep learning and the flexible numerical computation core is used across many other scientific domains. AI 38
- 39. Basic Classification in TensorFlow – Steps • Import the Fashion MNIST Dataset • Explore the Data • Preprocess the Data • Build the Model • Train the Model • Evaluate Accuracy • Make Predictions AI 39
- 40. Basic Classification in TensorFlow • Trains a neural network model to classify images of clothing, like sneakers and shirts. It's okay if you don't understand all the details, this is a fast-paced overview of a complete TensorFlow program with the details explained as we go. • This guide uses tf.keras, a high-level API to build and train models in TensorFlow. • Colab is the Resource provided by Google to write and test code on web. • https://colab.research.google.com/github/tensorflow/docs/blob/master/site/en/ tutorials/keras/basic_classification.ipynb#scrollTo=vasWnqRgy1H4 AI 40
- 42. Foundations Programming • Syntax and basic concepts: Google’s Python Class, Learn Python the Hard Way.Practice: Coderbyte, Codewars, HackerRank. Linear algebra • Deep Learning Book, Chapter 2: Linear Algebra. A quick review of the linear algebra concepts relevant to machine learning.A First Course in Linear Model Theory by Nalini Ravishanker and Dipak Dey. Textbook introducing linear algebra in a statistical context. Probability and Statistics • All of Statistics: A Concise Course in Statistical Inference, by Larry Wasserman. Introductory text on statistics. AI 42
- 43. Foundations Calculus • Khan Academy: Differential Calculus. Or, any introductory calculus course or textbook. Machine Learning • Andrew Ng’s Machine Learning course on Coursera Data science bootcamps: • Textbook - An Introduction to Statistical Learning by Gareth James et al. Excellent reference for essential machine learning concepts, available free online. Deep Learning • Deeplearning.ai, Andrew Ng’s introductory deep learning course.CS231n: Convolutional Neural Networks for Visual Recognition, Stanford’s deep learning course. Helpful for building foundations, with engaging lectures and illustrative problem sets. AI 43
- 44. References Deep Learning • Deeplearning.ai, Andrew Ng’s introductory deep learning course.CS231n: Convolutional Neural Networks for Visual Recognition, Stanford’s deep learning course. Helpful for building foundations, with engaging lectures and illustrative problem sets. AI 44
- 45. Questions AI 45