Continual Learning (CL) methods aim to learn from a sequence of tasks while avoiding the challenge of forgetting previous knowledge. We present DREAM-CL, a novel CL method for ECG arrhythmia detection that introduces dynamic prototype rehearsal memory. DREAM-CL selects representative prototypes by clustering data based on learning behavior during each training session. Within each cluster, we apply a smooth sorting operation that ranks samples by training difficulty, compressing extreme values and removing outliers. The more challenging samples are then chosen as prototypes for the rehearsal memory, ensuring effective knowledge retention across sessions. We evaluate our method on time-incremental, class-incremental, and lead-incremental scenarios using two widely used ECG arrhythmia datasets, Chapman and PTB-XL. The results demonstrate that DREAM-CL outperforms the state-of-the-art in CL for ECG arrhythmia detection. Detailed ablation and sensitivity studies are performed to validate the different design choices of our method.

Addressing the challenge of matching queries with the right experts amid temporal-textual inconsistencies, we present a novel approach that combines an attention-based text embedding model with a continuous-time module. This method effectively maps queries to relevant experts by analyzing concept-oriented vectors and user behavior, demonstrating significant effectiveness on StackOverflow and Yahoo datasets.

We propose a deep neural architecture that performs uncertainty-aware multi-view classification of arrhythmia from ECG. Our method learns two different views (1D and 2D) of single-lead ECG to capture different types of information. We use a fusion technique to reduce the conflict between the different views caused by noise and artifacts in ECG data, thus incorporating uncertainty to obtain stronger final predictions. Our framework contains the following three modules (1) a time-series module to learn the morphological features from ECG; (2) an image-space learning module to learn the spatiotemporal features; and (3) the uncertainty-aware fusion module to fuse the information from the two different views. Experimental results on two real-world datasets demonstrate that our framework not only improves the performance on arrhythmia classification compared to the state-of-the-art but also shows better robustness to noise and artifacts present in ECG.

Finding the most suitable individual to answer a question using brief content has important usages, including the community of question answering systems and online recommender frameworks. However, one must tackle challenges: Disregarding the indispensable noise in short text contents, authors usually answer the input query with mismatched words that can negatively influence the textual relevance. Moreover, many vocabularies imply various alterations. Finally, not every expert is eager to answer an input query given the time constraint, named the reluctance dilemma. To overcome the challenges, we devise a novel embedding approach that constructs context-aware vectors. We then extract the knowledge domains out of the online contextual content. While we track user textual-temporal behavioral patterns via an infinite continuous-time module, we recommend a set of experts pertinent to the given query and willingly provide the response during the expected time. Experimental results on two real-world datasets ofStackOverflowandYahooshow that our online time-sensitive value-wise transformer can achieve higher effectiveness and efficiency versus other trending rivals in online expert recommendation systems. In addition, we empirically experience that Fourier transformers can automatically infer multi-aspect base signals and overpass manual discrete-time models in obtaining time-specific user profiles.