Making Sense of Touch: Unsupervised Shapelet Learning in Bag-of-words Sense
This work addresses the challenge of interpreting tactile data in robotics, offering an incremental improvement in unsupervised feature learning for time-series analysis.
The paper tackles the problem of learning shapelets from long time-series data for robotics tasks by introducing NN-STNE, a neural network that uses t-SNE to map data into shapelet probabilities, resulting in improved clustering accuracy over state-of-the-art methods on the UCR dataset and an electrical component manipulation task.
This paper introduces NN-STNE, a neural network using t-distributed stochastic neighbor embedding (t-SNE) as a hidden layer to reduce input dimensions by mapping long time-series data into shapelet membership probabilities. A Gaussian kernel-based mean square error preserves local data structure, while K-means initializes shapelet candidates due to the non-convex optimization challenge. Unlike existing methods, our approach uses t-SNE to address crowding in low-dimensional space and applies L1-norm regularization to optimize shapelet length. Evaluations on the UCR dataset and an electrical component manipulation task, like switching on, demonstrate improved clustering accuracy over state-of-the-art feature-learning methods in robotics.