Class-incremental learning in the context of limited personal labeled samples (few-shot) is critical for numerous real-world applications, such as smart home devices. A key challenge in these scenarios is balancing the trade-off between adapting to new, personalized classes and maintaining the performance of the model on the original, base classes. Fine-tuning the model on novel classes often leads to the phenomenon of catastrophic forgetting, where the accuracy of base classes declines unpredictably and significantly. In this paper, we propose a simple yet effective mechanism to address this challenge by controlling the trade-off between novel and base class accuracy. We specifically target the ultra-low-shot scenario, where only a single example is available per novel class. Our approach introduces a Novel Class Detection (NCD) rule, which adjusts the degree of forgetting a priori while simultaneously enhancing performance on novel classes. We demonstrate the versatility of our solution by applying it to state-of-the-art Few-Shot Class-Incremental Learning (FSCIL) methods, showing consistent improvements across different settings. To better quantify the trade-off between novel and base class performance, we introduce new metrics: NCR@2FOR and NCR@5FOR. Our approach achieves up to a 30% improvement in novel class accuracy on the CIFAR100 dataset (1-shot, 1 novel class) while maintaining a controlled base class forgetting rate of 2%.

For smart grid services, accurate individual load forecasting is an essential element. When training individual forecasting models for multi-customers, discrepancies in data distribution among customers should be considered; there are two simple ways to build the models considering multi-customers: constructing each model independently or training as one model encompassing multi-customers. The independent approach shows higher accuracy than the latter. However, it deploys copious models, causing resource/management inefficiency; the latter is the opposite. A compromise between these two could be clustering-based forecasting. However, the previous studies are limited in applying to individual forecasting in that they focus on aggregated load and do not consider concept drift, which degrades accuracy over time. Therefore, we propose a distribution-aware temporal pooling framework that is enhanced clustering-based forecasting. For the clustering, we propose Variational Recurrent Deep Embedding (VaRDE) working in a distribution-aware manner, so it is suitable to process individual load. It allocates clusters to customers every time, so the clusters, where customers are assigned, are dynamically changed to resolve distribution change. We conducted experiments with real data for evaluation, and the result showed better performance than previous studies, especially with a few models even for unseen data, leading to high scalability.

As the deep learning (DL) has widely been used for application domains such as image classifications, natural language processing, and speech recognition, various software frameworks have been developed. They provide users with efficient programming interfaces for developing the DL applications. The optimization techniques within these frameworks generally are different from each other, which leads to different processing times for even the same applications. However, it is difficult that end users consider performance differences in processing time due to incompatible programming interface among the DL frameworks. These differences might cause redundant efforts and costs for end users to develop and maintain the applications. In this paper, we introduce an integrated deep learning engine (IDLE), a novel interface working on the top of the existing DL frameworks, which provides a convenient, flexible and scalable programming interface developing the DL applications for end users regardless of DL frameworks. Besides, we also propose a novel adaptive task scheduling scheme for training DL applications in a cluster with different GPUs. We implement our platform on the heterogeneous GPU cluster, and the results show that the proposed scheduling algorithm improves cost efficiency processing various DL applications.