Abstract: Complex factors make industrial power load forecasting difficult. Machine-learning methods are used to forecast industrial power load. Thus, a new power load forecasting method takes into account load characteristics in different regions, industries, and…
Abstract: Most head pose estimation methods require accurate ground-truth poses. The “ground truth” pose is often obtained subjectively by asking subjects to stare at different markers on the wall. Soft labels are better than explicit…
Abstract: Existing cross-domain recognition methods generally focus on changing data feature representation or classifier parameters, and their performance indicates their efficiency. Most methods do not integrate them into a single optimization objective to improve learning…
Abstract: Card fraud costs cardholders and issuers a lot. Modern methods detect fraud from transaction records using machine learning. Manually generating features requires domain knowledge and may be behind fraud’s modus operandi, so the online…
Abstract: Diverse ensembles are widely used in machine learning and credited with their success. Most of these methods increase ensemble diversity by sampling data or changing model structures. However, a family of ensemble models explicitly…
Abstract: Graph Neural Networks (GNNs) excel at recommendation, node classification, and link prediction. Graph neural network models create node embeddings by combining node features with neighboring node information. Most GNN models use a single aggregator…
Abstract: Machine learning’s fairness is a growing concern. Most machine-learning research is focused on either supervised or unsupervised learning. Two observations led us to consider semi-supervised learning for discrimination problems. First, previous research suggested that…
Abstract: One-hot encoding of categorical variables is often needed in statistical models. High-dimensional feature vectors make this strategy fail as categories increase. One-hot encoding also lacks morphological information for string entries. High-cardinality string categorical variables…
Abstract: The discriminative mixture variational autoencoder (DMVAE) is a deep probability model for semisupervised learning feature extraction. Encoding, decoding, and classification modules comprise the DMVAE. The encoder projects the observation to the latent space in…
Abstract: Spatial-temporal data is increasingly available due to the rapid development of GPS, mobile devices, and remote sensing. Human mobility understanding, smart transportation, urban planning, public safety, health care, and environmental management depend on spatio-temporal…
Abstract: Multimedia, machine learning, and data mining communities are focusing on multi-view clustering. Multi-view subspace clustering (MVSC) is a popular multi-view clustering algorithm because it can reveal the intrinsic low-dimensional clustering structure hidden across views….
Abstract: This article presents a data-driven method for computing reachable sets where active learning (AL) is used to reduce the computational burden. Set-based methods used to estimate reachable sets typically do not scale well with…
WhatsApp us