Henry Gouk's Home Page
I am a PhD student in the Machine Learning group at the University of Waikato. I am investigating how Deep Learning can be used create more robust visual SLAM systems. I am also interested in metric learning and various applications of it.
Gouk is a Scottish name, and is pronounced Gowk.
Software I have written can be found on my github profile. Most notable will be nnet, a CPU based Neural Network library that uses FFTs for fast convolutions.
- Cree, M. J., Perrone, J. A., Anthonys, G., Garnett, A. C., & Gouk, H. (2016, November). Estimating heading direction from monocular video sequences using biologically-based sensors. In Image and Vision Computing New Zealand (IVCNZ), 2016 International Conference on (pp. 1-6). IEEE. (PDF)
- Gouk, H., Pfahringer, B., & Cree, M. (2016, November). Learning Distance Metrics for Multi-Label Classification. In Proceedings of the 8th Asian Conference on Machine Learning. (PDF)
- Gouk, H., Pfahringer, B., & Cree, M. (2015). Fast Metric Learning For Deep Neural Networks. arXiv preprint arXiv:1511.06442. (PDF)
- Gouk, H. G. R., & Blake, A. M. (2014, November). Fast Sliding Window Classification with Convolutional Neural Networks. In Proceedings of the 29th International Conference on Image and Vision Computing New Zealand (p. 114). ACM. (PDF)
- Gouk, H. G. R. (2014 November). Accelerating Convolutional Neural Network Systems. (PDF)