Masking Ratio Variations in Time-Series Self-Supervised Learning: Convergence and Representation Quality

Self-supervised learning for time-series data holds potential similar to that recently unleashed in Natural Language Processing and Computer Vision. While most existing works in this area focus on contrastive learning, we propose a conceptually simple yet powerful non-contrastive approach, based on the data2vec self-distillation framework. The core of our method is a student-teacher scheme that predicts the latent representation of an input time series from masked views of the same time series. This strategy avoids strong modality-specific assumptions and biases typically introduced by the des

Research goal: What is the impact of masking ratio variations on the convergence speed and final representation quality of student-teacher schemes versus contrastive objectives in time-series self-supervised learning?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 8.1/10.