Time-series analysis is confounded by nonlinear time warping of the data. Traditional methods for joint alignment do not generalize: after aligning a given signal ensemble, they lack a mechanism, that does not require solving a new optimization problem, to align previously-unseen signals. In the multi-class case, they must also first classify the test data before aligning it. Here we propose the Diffeomorphic Temporal alignment Net (DTAN), a learning-based method for time-series joint alignment. Via flexible temporal transformer layers, DTAN learns and applies an input-dependent nonlinear time warping to its input signal. Once learned, DTAN easily aligns previously-unseen signals by its inexpensive forward pass. In a single-class case, the method is unsupervised: the ground-truth alignments are unknown. In the multi-class case, it is semi-supervised in the sense that class labels (but not the ground-truth alignments) are used during learning; in test time, however, the class labels are unknown. As we show, DTAN not only outperforms existing joint-alignment methods in aligning training data but also generalizes well to test data. Our code is available at https://github.com/BGU-CS-VIL/dtan.
Ron A Shapira Weber (Ben-Gurion University)
I am a Cognitive Science gradute student at Ben Gurion University. I am a part of both the Vision, Inference, and Learning (VIL) group, under the supervision of Dr. Oren Freifeld at the Computer Science Dept. and of the Computational Psychiatry Lab at the Dept. of Brain and Cognitive Science, under the supervision of Dr. Oren Shriki. My research areas are Machine Learning and Computational Neuroscience. Specificly, Deep Learning, Time-series Analysis and Computer Vision.
Matan Eyal (Ben Gurion University)
Nicki Skafte (Technical University of Denmark)
Oren Shriki (Ben-Gurion University of the Negev)
Oren Freifeld (Ben-Gurion University)
More from the Same Authors
2019 Poster: Reliable training and estimation of variance networks »
Nicki Skafte · Martin Jørgensen · Søren Hauberg
2019 Poster: Explicit Disentanglement of Appearance and Perspective in Generative Models »
Nicki Skafte · Søren Hauberg