Unsupervised Feature Learning from Temporal Data Article Swipe

PDF

Related Concepts

Pooling Computer science Feature learning Feature (linguistics) Artificial intelligence Unsupervised learning Encoder Focus (optics) Pattern recognition (psychology) Context (archaeology) Metric (unit) Autoencoder Machine learning Deep learning Economics Optics Physics Philosophy Operating system Biology Operations management Linguistics Paleontology

Ross Goroshin , Joan Bruna , Jonathan Tompson , David Eigen , Yann LeCun ·

YOU? · · 2015 · Open Access · · DOI: https://doi.org/10.48550/arxiv.1504.02518 · OA: W1699156674

Current state-of-the-art classification and detection algorithms rely on supervised training. In this work we study unsupervised feature learning in the context of temporally coherent video data. We focus on feature learning from unlabeled video data, using the assumption that adjacent video frames contain semantically similar information. This assumption is exploited to train a convolutional pooling auto-encoder regularized by slowness and sparsity. We establish a connection between slow feature learning to metric learning and show that the trained encoder can be used to define a more temporally and semantically coherent metric.