@conference{507,
  author = {Marelie Davel and Stefan Lotz and Marthinus Theunissen and Almaro De Villiers and Chara Grant and Randle Rabe and Cleo Conacher},
  title = {Knowledge Discovery in Time Series Data},
  abstract = {{\textbullet} Complex time series data often encountered in scientific and engineering domains.
{\textbullet} Deep learning (DL) is particularly successful here:
{\textendash} large data sets, multivariate input and/or ouput,
{\textendash} highly complex sequences of interactions.
{\textbullet} Model interpretability:
{\textendash} Ability to understand a model{\textquoteright}s decisions in a given context [1].
{\textendash} Techniques typically not originally developed for time series data.
{\textendash} Time series interpretations themselves become uninterpretable.
{\textbullet} Knowledge Discovery:
{\textendash} DL has potential to reveal interesting patterns in large data sets.
{\textendash} Potential to produce novel insights about the task itself [2, 3].
{\textbullet} {\textquoteleft}know-it{\textquoteright}: Collaborative project that studies knowledge discovery in
time series data.},
  year = {2023},
  journal = {Deep Learning Indaba 2023},
  month = {September 2023},
}