Introduction
1. Explain the concepts of Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL), and describe how they are interconnected.
- AI is an umbrella term that includes anything related to computer mimicking human intelligence.
- ML is a subset of AI that mainly refers to supervised and unsupervised learning, which builds models to learn, analyse patterns and structures and make decisions.
- DL is a subset of ML that builds models inspired by how human brain works (Adds layers in between input and output results)
2. Provide examples of supervised machine learning use-cases.
- Regression
- Classification
- Ranking
3. Provide examples of unsupervised machine learning use-cases.
- Clustering
- Dimensionality Reduction
- Anomaly Detection
4. How are supervised machine learning models trained?
Supervised ML models are trained on labeled data. In supervised ML, the learning process is “supervised” because it involves providing the model with a dataset that contains both input data and the corresponding correct output (or target) labels.
5. Explain the main use case of supervised machine learning models.
Supervised ML is used to learn the underlying patterns and relationships between the input data and the output labels so that it can make accurate predictions or classifications when given new, unseen data.
6. How are unsupervised machine learning models trained?
Unsupervised ML models are trained to learn from data without any explicit supervision or labeled outcomes.
7. Explain the main use case of unsupervised machine learning models.
Unsupervised ML is used to make sense of unstructured or unlabeled data on its own, seeking hidden patterns, structures, or relationships.
8. Explain the differences of evaluating unsupervised and supervised machine learning models.
- In supervised learning, model performance is evaluated by comparing its predictions to the true output labels in a testing dataset.
- Evaluating unsupervised learning models can be more subjective since there are often no “correct” answers to compare against.