Deep learning for time series classification Hassan Ismail

Deep learning for time series classification Hassan Ismail Fawaz Supervised by: Germain Forestier, Jonathan Weber, Lhassane Idoumghar, Pierre-Alain Muller

Time series are ubiquitous • Type of data present in numerous applicative domains • Allow to study the evolution in time of a process, a behavior, etc. [1] Bagnall, A. , Lines, J. , Bostrom, A. , Large, J. , & Keogh, E. (2017). The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances. Data Mining and Knowledge Discovery, 31(3), 606 -660.

Time series classification (TSC) • Time series are regrouped in classes (e. g. Gun / No Gun) • The goal is to affect a class to new time series [1] Bagnall, A. , Lines, J. , Bostrom, A. , Large, J. , & Keogh, E. (2017). The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances. Data Mining and Knowledge Discovery, 31(3), 606 -660.

Deep Convolutional Neural Network for TSC [1] Ismail Fawaz, H. , Forestier, G. , Weber, J. , Idoumghar, L. , & Muller, P. A. (2019). Deep learning for time series classification: a review. Data Mining and Knowledge Discovery, 33(4), 917 -963.

Convolution on images vs. time series Applying an edge detection convolution on an image Applying a learned discriminative convolution on the Gun. Point dataset [1] Ismail Fawaz, H. , Forestier, G. , Weber, J. , Idoumghar, L. , & Muller, P. A. (2019). Deep learning for time series classification: a review. Data Mining and Knowledge Discovery, 33(4), 917 -963.