A short survey on Interpretable techniques with Time series data

Artificial intelligence (AI) is a rapidly developing area that affects every aspect of modern life, from national security to hobbies like filmmaking and music production. However, because to its very human-engineered nature, AI has drawbacks, most notably the "black box" issue, which leaves decision-making procedures unclear. Because time series have many uses in the healthcare industry, this problem calls for the development of interpretability and explainability methodologies, especially for complicated deep learning models in time series analysis. This study compares and contrasts many time series interpretability techniques, assessing their advantages and disadvantages, and delving further into relevant context-specific interpretability issues.