Search

This article presents one-bit supervision, a novel setting of learning with fewer labels, for image classification. Instead of the training model using the accurate label of each sample, our setting requires the model to interact with the system by predicting the class label of each sample and learn from the answer whether the guess is correct, which provides one bit (yes or no) of information. An intriguing property of the setting is that the burden of annotation largely is alleviated in comparison to offering the accurate label. There are two keys to one-bit supervision: (i) improving the guess accuracy and (ii) making good use of the incorrect guesses. To achieve these goals, we propose a multi-stage training paradigm and incorporate negative label suppression into an off-the-shelf semi-supervised learning algorithm. Theoretical analysis shows that one-bit annotation is more efficient than full-bit annotation in most cases and gives the conditions of combining our approach with active learning. Inspired by this, we further integrate the one-bit supervision framework into the self-supervised learning algorithm, which yields an even more efficient training schedule. Different from training from scratch, when self-supervised learning is used for initialization, both hard example mining and class balance are verified to be effective in boosting the learning performance. However, these two frameworks still need full-bit labels in the initial stage. To cast off this burden, we utilize unsupervised domain adaptation to train the initial model and conduct pure one-bit annotations on the target dataset. In multiple benchmarks, the learning efficiency of the proposed approach surpasses that using full-bit, semi-supervised supervision. [ABSTRACT FROM AUTHOR]

Cell division cycle-associated (CDCA) gene family members are essential cell proliferation regulators and play critical roles in various cancers. However, the function of the CDCA family genes in gliomas remains unclear. This study aims to elucidate the role of CDCA family members in gliomas using in vitro and in vivo experiments and bioinformatic analyses. We included eight glioma cohorts in this study. An unsupervised clustering algorithm was used to identify novel CDCA gene family clusters. Then, we utilized multi-omics data to elucidate the prognostic disparities, biological functionalities, genomic alterations, and immune microenvironment among glioma patients. Subsequently, the scRNA-seq analysis and spatial transcriptomic sequencing analysis were carried out to explore the expression distribution of CDCA2 in glioma samples. In vivo and in vitro experiments were used to investigate the effects of CDCA2 on the viability, migration, and invasion of glioma cells. Finally, based on ten machine-learning algorithms, we constructed an artificial intelligence-driven CDCA gene family signature called the machine learning-based CDCA gene family score (MLCS). Our results suggested that patients with the higher expression levels of CDCA family genes had a worse prognosis, more activated RAS signaling pathways, and more activated immunosuppressive microenvironments. CDCA2 knockdown inhibited the proliferation, migration, and invasion of glioma cells. In addition, the MLCS had robust and favorable prognostic predictive ability and could predict the response to immunotherapy and chemotherapy drug sensitivity. • An integration analysis of CDCA gene family in gliomas by multiple omics data. • Novel CDCA family clusters with distinct signaling, genomic features, TIME, and prognosis. • Experiments showed CDCA2 acted as an oncogene in gliomas by RAS signaling. • An artificial intelligence-driven CDCA family signature was developed. • The MLCS was a stable and robust biomarker for prognosis and immunotherapy. [ABSTRACT FROM AUTHOR]