Temporal Sequence Modeling
Typical techniques for sequence modeling rely upon well-segmented sequences which have been edited to remove noisy or irrelevant parts. Therefore, we cannot easily apply such methods to noisy sequences expected in real-world applications.
In one of our projects, we study sequence modeling through the combination of RNNs that captures the temporal dependencies and the attention mechanism that localizes the salient observations which are relevant to the final decision and ignore the irrelevant (noisy) parts of the input sequence.
More recent work uses more powerful neural network models such as Transformers to process longer sequences. One of our more recent projects uses a hierarchical architecture to model multiple temporal resolutions of sequences, allowing representations of data with different degrees of granularity and more easily capturing long-range dependencies. This work is being applied to music generation, where modeling hierarchical structure is especially important.
Publications
W. Pei, T. Baltrušaitis, D. Tax and L.-P. Morency. Temporal Attention-Gated Model for Robust Sequence Classification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017
S. Rajagopalan, L.-P. Morency, T. Baltrus̆aitis and R. Goecke, Extending Long Short-Term Memory for Multi-View Structured Learning, In Proceedings of the European Conference on Computer Vision (ECCV), 2016
K. Bousmalis, S. Zafeiriou, L.-P. Morency, M. Pantic and Z. Ghahramani, Variational Infinite Hidden Conditional Random Fields, IEEE Transaction on Pattern Analysis and Machine Intelligence (TPAMI), Volume 37, Issue 9, September 2015, Pages 1917-1929
M. Ziaeefard, R. Bergevin and L.-P. Morency, Time-slice Prediction of Dyadic Human Activities. In Proceedings of the British Machine Vision conference (BMVC), 2015
C.-C. Chiu, L.-P. Morency and S. Marsella. Predicting Co-verbal Gestures: A Deep and Temporal Modeling Approach. In Proceedings of the International Conference on Intelligent Virtual Agents (IVA), 2015.
T. Baltrusaitis, P. Robinson and L.-P. Morency, Continuous Conditional Neural Fields for Structured Regression. In Proceedings of the 13th European Conference on Computer Vision (ECCV), 2014
K.Bousmalis, S. Zafeiriou, L.–P. Morency, M. Pantic and Z. Ghahramani, Variational Hidden Conditional Random Fields with Coupled Dirichlet Process Mixtures. In Proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases (ECML/PKDD), 2014
Y. Song, L.-P. Morency and R. Davis. Action Recognition by Hierarchical Sequence Summarization. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2013
J.-C. Levesque, C. Gagne and L.-P. Morency. Sequential Emotion Recognition using Latent-Dynamic Conditional Neural Fields. In Proceedings of the IEEE Conference on Automatic Face and Gesture Recognition (FG), 2013
K. Bousmalis, S. Zafeiriou, L.-P. Morency, M. Pantic and Z. Ghahramani. Variational Hidden Conditional Random Fields with Coupled Dirichlet Process Mixtures. In Proceedings of the European Conference on Machine Learning (ECML), 2013