The Machine Learning Landscape

The Machine Learning Landscape

1. What Is Machine Learning

   • The science of programming computers so they can learn from data.
   • Field of study that gives computers the ability to learn without being explicitly programmed.
   • A computer program is said to learn from experience "E" with respect to some task "T" and some performance measure "P", if its performance on "T", as measured by "P", improves with experience "E".

2. Why Use Machine Learning (ML is great for):

   • Problems for which existing solutions require a lot of fine-tuning or long lists of rules.
   • Complex problems for which using a traditional approach yields no good solution.
   • Fluctuating environments.
   • Getting insights about complex problems and large amounts of data.

3. Examples of Applications

   • Analyzing images of products on a production line to automatically classify them.
   • Detecting tumors in brain scans.
   • Automatically classifying news articles.
   • Automatically flagging offensive comments on discussion forums.
   • Summarizing long documents automatically.
   • Creating a chatbot or a personal assistant.
   • Forecasting your company´s revenue new year, based on many performance metrics.
   • Making your app react to voice commands.
   • Detecting credit card fraud.
   • Segmenting clients based on their purchases so that you can design a different marketing strategy for each segment.
   • Representing a complex, high-dimensional dataset in a clear and insightful diagram.
   • Recommending a product that a client may be interested in, based on past purchases.
   • Building an intelligent bot for a game

4. Types of Machine Learning Systems

   • How they are supervised during training
     • Supervised Learning
       • The training set you feed to the algorithm includes the desired solutions, called labels.
     • Unsupervised Learning
       • The training data is unlabeled. The system tries to learn without a teacher.
     • Semi-supervised Learning
       • The training data have plenty of unlabeled instances, and few labeled instances. In other words, the training data is partially labeled.
     • Self-supervised Learning
       • Involves generating a fully labeled dataset from a fully unlabeled one.
     • Reinforcement Learning
       • The learning system can observe the environment, select and perform actions and get rewards or penalties in return.
   • Whether or not they can learn incrementally on the fly
     • Batch Learning
       • The system is incapable of learning incrementally: it must be trained using all the available resources.
     • Online Learning
       • The system is training incrementally by feeding it data instances sequentially, either individually or in small groups called mini-batches.
   • Wheter they work by simply comparing new data points to known data points, or instead by detecting patterns in the training data and building a predictive model.
     • Instance-based Learning
       • The system learns the examples by heart, then generalizes to new cases by using a similarity measure to compare them to the learned examples.
     • Model-based Learning
       • From a set of examples, build a model of these examples and then use that model to make predictions.

5. Main Challenges of Machine Learning

   • Insufficient Quantity of Training Data
     • It takes a lot of data for most machine learning algorithms to work properly.
   • Nonrepresentative Training Data
     • By using a nonrepresentative training set, you trained a model that is unlikely to make accurate predictions.
   • Poor Quality Data
     • If the training data is full of errors, outliers and noise, it will make it harder for the system to detect the underlying patterns, so your system is less likely to perform well.
   • Irrelevant Features
     • The system will only be capable of learning if the training data contains enough relevant features and not too many irrelevant ones.
   • Overfitting the Training Data
     • It means that the model performs well on the training data, but it does not generalize well.
   • Underfitting the Training Data
     • It occurs when your model is too simple to learn the underlying structure of the data.

6. Testing and Validating

   • Hyperparameter Tuning and Model Selection
     • Model Selection
       • When deciding between different types of models (e.g., linear vs. polynomial), you typically train both models and compare their generalization performance using a test set.
     • Regularization Hyperparameter Tuning
       • Choosing the right value for the regularization hyperparameter involves training multiple models with different hyperparameter values and selecting the one with the lowest generalization error.
     • Holdout Validation
       • This involves splitting the training set into a smaller training set and a validation set. Multiple models with different hyperparameters are trained on the reduced training set, and the best-performing model is selected based on its performance on the validation set. The final model is then trained on the full training set (including the validation set) and evaluated on the test set.
     • Repeated Cross-Validation
       • To mitigate the potential drawbacks of holdout validation. repeated cross-validation can be employed. This involves splitting the training set into multiple subsets (folds), training the model on all but one fold, and evaluating it on the remaining fold. This process is repeated multiple times with different fold combinations, and the performance of the model is averaged across all evaluations.