SlideShare a Scribd company logo

Machine learning: A Walk Through School Exams

•
•
Ramsha Ijaz
Ramsha Ijaz

When it comes to studying, Machines and Students have one thing in common: Examinations. To perform well on their final evaluations, humans require taking classes, reading books and solving practice quizzes. Similarly, machines need artificial intelligence to memorize data, infer feature correlations, and pass validation standards in order to solve almost any problem. In this quick introductory session, we'll walk through these analogies to learn the core concepts behind Machine Learning, and why it works so well!

Related slideshows

RecSys 2016 Talk: Feature Selection For Human Recommenders
RecSys 2016 Talk: Feature Selection For Human RecommendersRecSys 2016 Talk: Feature Selection For Human Recommenders
RecSys 2016 Talk: Feature Selection For Human Recommenders
 
Supervised Machine Learning
Supervised Machine LearningSupervised Machine Learning
Supervised Machine Learning
 
Machine learning overview
Machine learning overviewMachine learning overview
Machine learning overview
 
1 of 36
Machine Learning and A Walk through School Exams! Introduction to Artificial Intelligence
SPEAKER IMAGE Ramsha Siddiqui Associate Data Scientist (NLP) i2c inc. Lahore, Pakistan ● 💬 Conversational AI and Dialog Systems. ● 👩💻 Google Developers Group, Lahore. ● 🥇 WTM and WDA Scholar, 2018-2020. ● 🎓 FAST NUCES, Lahore - Spring, 2019. ● ✈️ Global UGRAD Alumnus - UIW, Texas. ● 💼 Co-Founder of Startup - FunKadaa.
What is Machine Learning? Machine Learning is the application of Artificial Intelligence that allows machines to learn and improve from experience, without being explicitly programmed, to solve almost any problem in the world.
How do humans learn?
The Life of a Student 1. Gather Knowledge
The Life of a Student 1. Gather Knowledge 2. Extract Concepts
The Life of a Student 2. Extract Concepts 3. Practice Concepts 1. Gather Knowledge
The Life of a Student 3. Practice Concepts 4. Test Yourself 1. Gather Knowledge 2. Extract Concepts
Learning Chess! 3. Practice 4. Compete! 1. Learn Pieces and Moves 2. Understand & Memorize
How do machines learn?
1. Gather Training Data The initial set of data-points required to learn about a problem, in order to solve it.
2. Data Preprocessing Analyzing Data and Extracting or Deriving useful features from it, in efforts to improve results.
Why is it important?
3. Model Selection / Construction An algorithm / mathematical derivation containing variables that can be derived / learned from Training Data is called a Machine Learning Model. The values of the variables are saved after training to be used for predictions later.
3. Model Selection / Construction Selecting the right “kind of model” for your Machine Learning is also important, because feature-learning / data-understanding is an internal process for machines, and certain types of problems are more well-suited for certain types of algorithms. Let’s look at the two broadest types!
Supervised Learning When a Model’s Expected Outputs are defined, it’s called a Supervised Learning Problem. It mainly has two sub-categories: ● Classification (categorizing input data into classes) ● Regression (predicting a continuous / real value instead of an output class-label)
Example: Facial Recognition Recognizing people’s faces and its features (eyes, head-pose, etc.)
Example: Facial Recognition Convolutional Neural Networks (CNNs) have been classically useful for the task of identifying different components / features of images.
Example: Speech Recognition Converting Audio Sequences (Speech) into Text Sequences.
Example: Speech Recognition Sequence to Sequence Networks (Seq2Seq) have so far shown promising results for this task (with updated research and architectures being released almost every day!)
Example: Covid Projection Predicting the number of Covid-Cases every other day / month.
Example: Covid Projection Time-Series Forecasting is a common problem in Machine Learning, and all models used for that task, can be trained here for comparison.
A Note: Some of these problems such as ● Image Classification ● Text Classification, etc. you’ll find that some Model Architectures are also problem- favorable, as they’ve shown good results before. It’s always wise for you to do some Research beforehand to find good Model Implementations / Datasets on the task you’re working on, before starting from scratch.
Unsupervised Learning When a Model’s Expected Outputs are undefined, and input-data only needs to be clustered or segmented into somewhat meaningful groups, it’s called Unsupervised Learning. Example: ● K-Means Algorithm (Finds Cluster Centroids based on the no. of centers defined = K)
Example: Topic Modeling Training a model to learn the topic of a document based on the correlations of different words used inside it, without knowing prior what those topics are.
A Note: Know your Problem Domain!
4. Model Training Feeding your Input Data to your Model for learning - is called Model Training. Here’s some things you may wanna do before that: ● Data Folds: Splitting your Training Data into two sets: One for Training on, and one for Validating Performance. ● Epochs: The number of times your model sees the Training Data (revising it). ● Model Parameters: Take out a set of possible parameter values that you want to tweak in your model.
Example: w1*x1 + w2*x2 + w3*x3 = y where for every example, we have input-values “x” and model-variables “w”.
5. Model Testing Alas, models can’t escape examinations either. After Training a Model, we want to see Model Predictions, and compare the values that model got with those it got incorrect, and evaluate overall performance. Choosing the right performance metric matters and varies per problem!
Task Performance
On average, most humans and models are able to learn almost any task following this method. In terms of performance comparisons however, sometimes we: ● Underachieve ● Overachieve
Underachieving Reasons: ● Unseen Test Data: If the data in the examination is totally unrelated to the Train Data, then the model will get confused. Evenly distribute your data between Train and Test Sets. ● Insufficient Training Data: Perhaps the amount of Training Data is insufficient / too difficult to learn anything useful from. Add more data or try new features. ● Model Architecture: The Model selected for a particular task may not be the right choice for it.
Overachieving Reasons: ● Randomness: Sometimes it’s just dumb luck. (Models are initialized with Random Variable Values). ● Model Pretraining: If your Model was trained on a task similar to this one before and has a good memory, it may perform well even if your Training Data was pretty small / useless. A lot of research is going into pre-training these days!
Model Predictions In order to trust that our model’s are working correctly, it’s always a good idea to provide additional explainability with the model output like: ● Confidence Score(s) ● Input Features
Let’s quickly Train a Model: https://teachablemachine.withgoogle.com/
Thank you! Any questions? Find me: ramsha@gdglahore.com

More Related Content

Machine learning: A Walk Through School Exams

  • 1. Machine Learning and A Walk through School Exams! Introduction to Artificial Intelligence
  • 2. SPEAKER IMAGE Ramsha Siddiqui Associate Data Scientist (NLP) i2c inc. Lahore, Pakistan ● 💬 Conversational AI and Dialog Systems. ● 👩💻 Google Developers Group, Lahore. ● 🥇 WTM and WDA Scholar, 2018-2020. ● 🎓 FAST NUCES, Lahore - Spring, 2019. ● ✈️ Global UGRAD Alumnus - UIW, Texas. ● 💼 Co-Founder of Startup - FunKadaa.
  • 3. What is Machine Learning? Machine Learning is the application of Artificial Intelligence that allows machines to learn and improve from experience, without being explicitly programmed, to solve almost any problem in the world.
  • 4. How do humans learn?
  • 5. The Life of a Student 1. Gather Knowledge
  • 6. The Life of a Student 1. Gather Knowledge 2. Extract Concepts
  • 7. The Life of a Student 2. Extract Concepts 3. Practice Concepts 1. Gather Knowledge
  • 8. The Life of a Student 3. Practice Concepts 4. Test Yourself 1. Gather Knowledge 2. Extract Concepts
  • 9. Learning Chess! 3. Practice 4. Compete! 1. Learn Pieces and Moves 2. Understand & Memorize
  • 10. How do machines learn?
  • 11. 1. Gather Training Data The initial set of data-points required to learn about a problem, in order to solve it.
  • 12. 2. Data Preprocessing Analyzing Data and Extracting or Deriving useful features from it, in efforts to improve results.
  • 13. Why is it important?
  • 14. 3. Model Selection / Construction An algorithm / mathematical derivation containing variables that can be derived / learned from Training Data is called a Machine Learning Model. The values of the variables are saved after training to be used for predictions later.
  • 15. 3. Model Selection / Construction Selecting the right “kind of model” for your Machine Learning is also important, because feature-learning / data-understanding is an internal process for machines, and certain types of problems are more well-suited for certain types of algorithms. Let’s look at the two broadest types!
  • 16. Supervised Learning When a Model’s Expected Outputs are defined, it’s called a Supervised Learning Problem. It mainly has two sub-categories: ● Classification (categorizing input data into classes) ● Regression (predicting a continuous / real value instead of an output class-label)
  • 17. Example: Facial Recognition Recognizing people’s faces and its features (eyes, head-pose, etc.)
  • 18. Example: Facial Recognition Convolutional Neural Networks (CNNs) have been classically useful for the task of identifying different components / features of images.
  • 19. Example: Speech Recognition Converting Audio Sequences (Speech) into Text Sequences.
  • 20. Example: Speech Recognition Sequence to Sequence Networks (Seq2Seq) have so far shown promising results for this task (with updated research and architectures being released almost every day!)
  • 21. Example: Covid Projection Predicting the number of Covid-Cases every other day / month.
  • 22. Example: Covid Projection Time-Series Forecasting is a common problem in Machine Learning, and all models used for that task, can be trained here for comparison.
  • 23. A Note: Some of these problems such as ● Image Classification ● Text Classification, etc. you’ll find that some Model Architectures are also problem- favorable, as they’ve shown good results before. It’s always wise for you to do some Research beforehand to find good Model Implementations / Datasets on the task you’re working on, before starting from scratch.
  • 24. Unsupervised Learning When a Model’s Expected Outputs are undefined, and input-data only needs to be clustered or segmented into somewhat meaningful groups, it’s called Unsupervised Learning. Example: ● K-Means Algorithm (Finds Cluster Centroids based on the no. of centers defined = K)
  • 25. Example: Topic Modeling Training a model to learn the topic of a document based on the correlations of different words used inside it, without knowing prior what those topics are.
  • 26. A Note: Know your Problem Domain!
  • 27. 4. Model Training Feeding your Input Data to your Model for learning - is called Model Training. Here’s some things you may wanna do before that: ● Data Folds: Splitting your Training Data into two sets: One for Training on, and one for Validating Performance. ● Epochs: The number of times your model sees the Training Data (revising it). ● Model Parameters: Take out a set of possible parameter values that you want to tweak in your model.
  • 28. Example: w1*x1 + w2*x2 + w3*x3 = y where for every example, we have input-values “x” and model-variables “w”.
  • 29. 5. Model Testing Alas, models can’t escape examinations either. After Training a Model, we want to see Model Predictions, and compare the values that model got with those it got incorrect, and evaluate overall performance. Choosing the right performance metric matters and varies per problem!
  • 31. On average, most humans and models are able to learn almost any task following this method. In terms of performance comparisons however, sometimes we: ● Underachieve ● Overachieve
  • 32. Underachieving Reasons: ● Unseen Test Data: If the data in the examination is totally unrelated to the Train Data, then the model will get confused. Evenly distribute your data between Train and Test Sets. ● Insufficient Training Data: Perhaps the amount of Training Data is insufficient / too difficult to learn anything useful from. Add more data or try new features. ● Model Architecture: The Model selected for a particular task may not be the right choice for it.
  • 33. Overachieving Reasons: ● Randomness: Sometimes it’s just dumb luck. (Models are initialized with Random Variable Values). ● Model Pretraining: If your Model was trained on a task similar to this one before and has a good memory, it may perform well even if your Training Data was pretty small / useless. A lot of research is going into pre-training these days!
  • 34. Model Predictions In order to trust that our model’s are working correctly, it’s always a good idea to provide additional explainability with the model output like: ● Confidence Score(s) ● Input Features
  • 35. Let’s quickly Train a Model: https://teachablemachine.withgoogle.com/
  • 36. Thank you! Any questions? Find me: ramsha@gdglahore.com