Learning Types Flashcards
1
Q
What are the main supervised learning tasks?
A
Classification and regression.
2
Q
What are examples of supervised learning algorithms?
A
- k-NN
- Linear Regression
- Logistic Regression
- SVM
- Decision Trees
- Random Forest
- NN
3
Q
What are the common unsupervised learning tasks?
A
- Clustering
- Anomaly Detection
- Visualization and Dimensionality Reduction
- Association Rule Learning
4
Q
What are some of the algorithms used in Clustering?
A
- K-Means
- DB-Scan
- Hierarchical Clustering Analysis
5
Q
What are some algorithms used for Anomaly Detection?
A
- One-Class SVM
- Isolation Forest
6
Q
What are some algorithms used for Visualization and Dimensionality Reduction?
A
- Principle Component Analysis
- Kernel PCA
- Locally Linear Embedding
- t-Distributed Stochastic Neighbour Embedding
7
Q
What are some algorithms used for Association Rule Learning?
A
- Apriori
- Eclat
8
Q
What are the advantages to Online Learning?
A
- Learning steps are faster and cheaper
- Systems can independently adapt on the fly
- Handles a continuous dataflow well
- Memory efficient (used examples are discarded)
- Useful for datasets too large to fit in memory
9
Q
What are the challenges with Online Learning?
A
- Setting the learning rate
- Large amounts of bad data will destroy system performance
10
Q
What is the trade-off associated with the Learning Rate in Online learning?
A
- Faster rates will learn new data quickly, but forget old data quickly too
- Slower rates will learn new data slowly, but forget old data slowly too