Skip to main content

Table 2 Comparison among Different machine learning methods

From: Cryptocurrency trading: a comprehensive survey

 

Supervised learning

Unsupervised learning

Reinforcement learning

Definition

The machine learns by using labeled data

Machine training through unlabelled data without any guidance

Agents interact with their environment by performing actions and learning from errors or rewards

Type of problems

Rgression or classification

Association or clustering

Reward-based

Type of data

Labeled adata

Unlabeled data

No predefined data

Training

External supervision

No supervision

No supervision

Approach

Mapping tagged inputs to unknown outputs

Understanding patterns or finding outputs

Follow the trail-and-error method