Unsupervised representation learning approaches aim to learn discriminative feature
representations from unlabeled data, without the requirement of annotating every sample …

In real-world scenarios, achieving domain generalization (DG) presents significant
challenges as models are required to generalize to unknown target distributions …

Machine learning models often fail to generalize well under distributional shifts.
Understanding and overcoming these failures have led to a research field of Out-of …

Domain generalization (DG) approaches typically use the hypothesis learned on source
domains for inference on the unseen target domain. However, such a hypothesis can be …

Abstract In sensor-based Human Activity Recognition (HAR), a predominant assumption is
that the data utilized for training and evaluation purposes are drawn from the same …

Foundation models, especially LLMs, are profoundly transforming deep learning. Instead of
training many task-specific models, we can adapt a single pretrained model to many tasks …

Although pre-trained transformers and reprogrammed text-based LLMs have shown strong
performance on time series tasks, the best-performing architectures vary widely across …

Human activity recognition (HAR) is a time series classification task that focuses on
identifying the motion patterns from human sensor readings. Adequate data is essential but …

Time Series Representation Learning (TSRL) focuses on generating informative
representations for various Time Series (TS) modeling tasks. Traditional Self-Supervised …

Invariance learning methods aim to learn invariant features in the hope that they generalize
under distributional shifts. Although many tasks are naturally characterized by continuous …