Wormhole: Concept-Aware Deep Representation Learning for Co-Evolving Sequences

TL;DR

Wormhole is a deep learning framework that identifies and tracks dynamic concepts in co-evolving sequences, improving interpretability and detection of concept shifts in complex temporal data.

Contribution

It introduces a concept-aware representation model with a self-representation layer and a wormhole mechanism for detecting concept transitions in time sequences.

Findings

Effective segmentation of time series into meaningful concepts

Accurate detection of concept transitions and drifts

Enhanced interpretability of temporal representations

Abstract

Identifying and understanding dynamic concepts in co-evolving sequences is crucial for analyzing complex systems such as IoT applications, financial markets, and online activity logs. These concepts provide valuable insights into the underlying structures and behaviors of sequential data, enabling better decision-making and forecasting. This paper introduces Wormhole, a novel deep representation learning framework that is concept-aware and designed for co-evolving time sequences. Our model presents a self-representation layer and a temporal smoothness constraint to ensure robust identification of dynamic concepts and their transitions. Additionally, concept transitions are detected by identifying abrupt changes in the latent space, signifying a shift to new behavior - akin to passing through a wormhole. This novel mechanism accurately discerns concepts within co-evolving sequences and pinpoints the exact locations of these wormholes, enhancing the interpretability of the learned representations. Experiments demonstrate that this method can effectively segment time series data into meaningful concepts, providing a valuable tool for analyzing complex temporal patterns and advancing the detection of concept drifts.