Human Activity Recognition from Wearable Sensor Data Using Self-Attention
- Saif Mahmud, M. T. H. Tonmoy, A. Ali
- Computer ScienceEuropean Conference on Artificial Intelligence
- 17 March 2020
This work proposes a self-attention based neural network model that foregoes recurrent architectures and utilizes different types of attention mechanisms to generate higher dimensional feature representation used for classification in human activity recognition.
GRU-based Attention Mechanism for Human Activity Recognition
- Md. Nazmul Haque, M. Tanjid Hasan Tonmoy, Saif Mahmud, A. Ali, Muhammad Asif Hossain Khan, M. Shoyaib
- Computer Science1st International Conference on Advances in…
- 1 May 2019
The introduced model has achieved better performance with respect to the well-defined evaluation metrics in both uniform and imbalanced class distribution than the existing state-of-the-art deep learning based model.
Hierarchical Self Attention Based Autoencoder for Open-Set Human Activity Recognition
- M. T. H. Tonmoy, Saif Mahmud, A. K. M. M. Rahman, M. Amin, A. Ali
- Computer SciencePacific-Asia Conference on Knowledge Discovery…
- 7 March 2021
The proposed self attention based approach combines data hierarchically from different sensor placements across time to classify closed-set activities and it obtains notable performance improvement over state-of-the-art models on five publicly available datasets.
STLGRU: Spatio-Temporal Lightweight Graph GRU for Traffic Flow Prediction
- Kishor Kumar Bhaumik, Fahim Faisal Niloy, Saif Mahmud, Simon S. Woo
- Computer ScienceArXiv
- 8 December 2022
This paper proposes a novel deep learning framework - STLGRU that can effectively capture both local and global spatial-temporal relations of a traffic network using memory-augmented attention and gating mechanism and shows that the model performs better than existing methods while the memory footprint remains lower.