Weakly Supervised AUC Optimization: A Unified Partial AUC Approach.

Since acquiring perfect supervision is usually difficult, real-world machine learning tasks often confront inaccurate, incomplete, or inexact supervision, collectively referred to as weak supervision. In this work, we present WSAUC, a unified framework for weakly supervised AUC optimization problems, which covers noisy label learning, positive-unlabeled learning, multi-instance learning, and semi-supervised learning scenarios. Within the WSAUC framework, we first frame the AUC optimization problems in various weakly supervised scenarios as a common formulation of minimizing the AUC risk on contaminated sets, and demonstrate that the empirical risk minimization problems are consistent with the true AUC. Then, we introduce a new type of partial AUC, specifically, the reversed partial AUC (rpAUC), which serves as a robust training objective for AUC maximization in the presence of contaminated labels. WSAUC offers a universal solution for AUC optimization in various weakly supervised scenarios by maximizing the empirical rpAUC. Theoretical and experimental results under multiple settings support the effectiveness of WSAUC on a range of weakly supervised AUC optimization tasks.

Theoretical Exploration of Flexible Transmitter Model.

Neural network models generally involve two important components, i.e., network architecture and neuron model. Although there are abundant studies about network architectures, only a few neuron models have been developed, such as the MP neuron model developed in 1943 and the spiking neuron model developed in the 1950s. Recently, a new bio-plausible neuron model, flexible transmitter (FT) model (Zhang and Zhou, 2021), has been proposed. It exhibits promising behaviors, particularly on temporal-spatial signals, even when simply embedded into the common feedforward network architecture. This article attempts to understand the properties of the FT network (FTNet) theoretically. Under mild assumptions, we show that: 1) FTNet is a universal approximator; 2) the approximation complexity of FTNet can be exponentially smaller than those of commonly used real-valued neural networks with feedforward/recurrent architectures and is of the same order in the worst case; and 3) any local minimum of FTNet is the global minimum, implying that it is possible to identify global minima by local search algorithms.

Optical coherence tomography in patients with Wilson's disease.

BACKGROUND: Morphological changes of retina in patients with Wilson's disease (WD) can be found by optical coherence tomography (OCT), and such changes had significant differences between neurological forms (NWD) and hepatic forms (HWD) of WD. The aim of this study was to evaluate the relationship between morphological parameters of retina and brain magnetic resonance imaging (MRI) lesions, course of disease, type of disease, and sexuality in WD. METHODS: A total of 46 WD patients and 40 health controls (HC) were recruited in this study. A total of 42 WD patients were divided into different groups according to clinical manifestations, course of disease, sexuality, and brain MRI lesions. We employed the Global Assessment Scale to assess neurological severity of WD patients. All WD patients and HC underwent retinal OCT to assess the thickness of inner limiting membrane (ILM) layer to retinal pigment epithelium layer and inner retina layer (ILM to inner plexiform layer, ILM-IPL). RESULTS: Compared to HWD, NWD had thinner superior parafovea zone (108.07 ± 6.89 vs. 114.40 ± 5.54 µm, p < .01), temporal parafovea zone (97.17 ± 6.65 vs. 103.60 ± 4.53 µm, p < .01), inferior parafovea zone (108.114 ± 7.65 vs. 114.93 ± 5.84 µm, p < .01), and nasal parafovea zone (105.53 ± 8.01 vs. 112.10 ± 5.44 µm, p < .01) in inner retina layer. Course of disease influenced the retina thickness. Male patients had thinner inner retina layer compared to female patients. CONCLUSION: Our results demonstrated that WD had thinner inner retina layer compared to HC, and NWD had thinner inner retina layer compared to HWD. We speculated the thickness of inner retina layer may be a potential useful biomarker for NWD.

Analysis of risk factors for neurological symptoms in patients with purely hepatic Wilson's disease at diagnosis.

OBJECTIVE: To analyze and explore the risk factors for neurological symptoms in patients with purely hepatic Wilson's disease (WD) at diagnosis. METHODS: This retrospective study was conducted at the First Affiliated Hospital of the Guangdong Pharmaceutical University on 68 patients with purely hepatic WD aged 20.6 ± 7.2 years. The physical examinations, laboratory tests, color Doppler ultrasound of the liver and spleen, and magnetic resonance imaging (MRI) of the brain were performed. RESULTS: The elevated alanine transaminase (ALT) and aspartate transaminase (AST) levels and 24-h urinary copper level were higher in the purely hepatic WD who developed neurological symptoms (NH-WD) group than those in the purely hepatic WD (H-WD) group. Adherence to low-copper diet, and daily oral doses of penicillamine (PCA) and zinc gluconate (ZG) were lower in the NH-WD group than those in the H-WD group. Logistic regression analysis showed that insufficient doses of PCA and ZG were associated with the development of neurological symptoms in patients with purely hepatic WD at diagnosis. CONCLUSION: The development of neurological symptoms in patients with purely hepatic WD was closely associated with insufficient doses of PCA and ZG, and the inferior efficacy of copper-chelating agents. During the course of anti-copper treatment, the patient's medical status and the efficacy of copper excretion should be closely monitored.

Elderly onset of MELAS in a male: A case report.

Background: Mitochondrial encephalomyopathy with lactic acidosis and stroke-like episodes (MELAS) is one of the most common maternally inherited mitochondrial diseases which rarely affects elderly people. Case presentation: We reported the case of a 61-year-old male patient with MELAS. He was experiencing acute migraine-like headaches as the first symptoms. Laboratory data showed elevated lactate and creatine kinase levels. Brain magnetic resonance imaging (MRI) found a high signal intensity lesion in the left occipital-temporal-parietal lobe on diffusion-weighted imaging (DWI). Magnetic resonance angiography (MRA) revealed reversible vasoconstriction of the middle cerebral arteries and superficial temporal arteries. A muscle biopsy suggested minor muscle damage. A genetic study revealed a mitochondrial DNA A3243G mutation. Conclusion: Elderly onset of MELAS is rare and easily misdiagnosed as an ischemic stroke. MELAS with the onset of stroke-like episodes should be considered in adult or elderly patients with imaging findings that are atypical for cerebral infarction. The use of multimodal MRI in the clinical diagnosis of MELAS could be extremely beneficial.

Aqueous C-H aminomethylation of phenols by iodine catalysis.

A transition-metal-free strategy regarding an iodine-sodium percarbonate catalysis to achieve the ortho-aminomethylation of phenols in aqueous media has been developed. This method can effectively broaden a wide range of phenols, tolerate sensitive functional groups, and achieve the late-stage functionalization of ten functional molecules that contain phenolic structures.

Linear structural features of Wilson's disease and its correlation with neurological symptoms.

To measure the linear structure of the brain in patients with Wilson's disease (WD) and analyze its correlation with neurological symptoms. A total of 174 patients diagnosed with WD were enrolled. According to the type of clinical presentation, the patients with WD were divided into two groups: neurological (NWD) and hepatic (HWD). Sixty healthy volunteers were assigned to a control group. All patients with WD and healthy controls underwent brain magnetic resonance imaging (MRI). The severity of the neurological symptoms was assessed using the Burke Fahn Marsden Movement subscale (BFM-M). Linear brain measurements were performed using T1-weighted MRI scans of all the patients, and the correlation between these linear indices and BFM-M score was investigated. The Huckman index, third ventricle width, and sulcus width of the NWD group were significantly higher than those of the HWD and control groups (P < .05). The frontal horn index, ventricular index, and lateral ventricular body width index of the NWD group were significantly lower than those of the HWD and control groups (P < .05). The Huckman index and third ventricle width of the HWD group were higher than those of the control group (P < .05), whereas the body width index of the lateral ventricle was lower than that of the control group (P < .05). The BFM-M score correlated with the Huckman index (r = 0.29, P < .05), third ventricle width (r = 0.426, P < .001), and lateral ventricular body width index (r = -0.19, P < .05). This study demonstrated significant changes in the linear structure of patients with WD. Linear brain measurement analysis could be used as a potential method to assess the severity of neurological symptoms in WD.

Open-environment machine learning.

Conventional machine learning studies generally assume close-environment scenarios where important factors of the learning process hold invariant. With the great success of machine learning, nowadays, more and more practical tasks, particularly those involving open-environment scenarios where important factors are subject to change, called open-environment machine learning in this article, are present to the community. Evidently, it is a grand challenge for machine learning turning from close environment to open environment. It becomes even more challenging since, in various big data tasks, data are usually accumulated with time, like streams, while it is hard to train the machine learning model after collecting all data as in conventional studies. This article briefly introduces some advances in this line of research, focusing on techniques concerning emerging new classes, decremental/incremental features, changing data distributions and varied learning objectives, and discusses some theoretical issues.

A patient with hereditary transthyretin amyloidosis involving multiple cranial nerves due to a rare p.(Phe84Ser) variant.

We report a 30-year-old man involving gastrointestinal symptoms, vitreous opacity, and multiple cranial neuropathies. Transthyretin-related hereditary amyloidosis genetic testing revealed a rare c.251T > C variant p.(Phe84Ser). Only four cases with this variant have been reported before.

Improving generalization of deep neural networks by leveraging margin distribution.

Recent research has used margin theory to analyze the generalization performance for deep neural networks (DNNs). The existed results are almost based on the spectrally-normalized minimum margin. However, optimizing the minimum margin ignores a mass of information about the entire margin distribution, which is crucial to generalization performance. In this paper, we prove a generalization upper bound dominated by the statistics of the entire margin distribution. Compared with the minimum margin bounds, our bound highlights an important measure for controlling the complexity, which is the ratio of the margin standard deviation to the expected margin. We utilize a convex margin distribution loss function on the deep neural networks to validate our theoretical results by optimizing the margin ratio. Experiments and visualizations confirm the effectiveness of our approach and the correlation between generalization gap and margin ratio.

Towards understanding theoretical advantages of complex-reaction networks.

Complex-valued neural networks have attracted increasing attention in recent years, while it remains open on the advantages of complex-valued neural networks in comparison with real-valued networks. This work takes one step on this direction by introducing the complex-reaction network with fully-connected feed-forward architecture. We prove the universal approximation property for complex-reaction networks, and show that a class of radial functions can be approximated by a complex-reaction network using the polynomial number of parameters, whereas real-valued networks need at least exponential parameters to reach the same approximation level. For empirical risk minimization, we study the landscape and convergence of complex gradient descents. Our theoretical result shows that the critical point set of complex-reaction networks is a proper subset of that of real-valued networks, which may show some insights on finding the optimal solutions more easily for complex-reaction networks.

Prediction With Unpredictable Feature Evolution.

Learning with feature evolution studies the scenario where the features of the data streams can evolve, i.e., old features vanish and new features emerge. Its goal is to keep the model always performing well even when the features happen to evolve. To tackle this problem, canonical methods assume that the old features will vanish simultaneously and the new features themselves will emerge simultaneously as well. They also assume that there is an overlapping period where old and new features both exist when the feature space starts to change. However, in reality, the feature evolution could be unpredictable, which means that the features can vanish or emerge arbitrarily, causing the overlapping period incomplete. In this article, we propose a novel paradigm: prediction with unpredictable feature evolution (PUFE) where the feature evolution is unpredictable. To address this problem, we fill the incomplete overlapping period and formulate it as a new matrix completion problem. We give a theoretical bound on the least number of observed entries to make the overlapping period intact. With this intact overlapping period, we leverage an ensemble method to take the advantage of both the old and new feature spaces without manually deciding which base models should be incorporated. Theoretical and experimental results validate that our method can always follow the best base models and, thus, realize the goal of learning with feature evolution.

Margin Distribution Analysis.

Margin is an important concept in machine learning; theoretical analyses further reveal that the distribution of margin plays a more critical role than the minimum margin in generalization power. Recently, several approaches have achieved performance breakthroughs by optimizing the margin distribution, but their computational cost, which is usually higher than before, still hinders them to be widely applied. In this article, we propose margin distribution analysis (MDA), which optimizes the margin distribution more simply by maximizing the margin mean and minimizing the margin variance simultaneously. MDA is efficient and resistive to class-imbalance naturally, since its objective distinguishes the margin means of different classes and can be broken up into two linear equations. In practice, it can also cooperate with other frameworks such as reweight-minimization when facing complex circumstances with noise and outliers. Empirical studies validate the superiority of MDA in real-world data sets, and demonstrate that simple approaches can also perform competitively by optimizing margin distribution.

Flexible Transmitter Network.

Current neural networks are mostly built on the MP model, which usually formulates the neuron as executing an activation function on the real-valued weighted aggregation of signals received from other neurons. This letter proposes the flexible transmitter (FT) model, a novel bio-plausible neuron model with flexible synaptic plasticity. The FT model employs a pair of parameters to model the neurotransmitters between neurons and puts up a neuron-exclusive variable to record the regulated neurotrophin density. Thus, the FT model can be formulated as a two-variable, two-valued function, taking the commonly used MP neuron model as its particular case. This modeling manner makes the FT model biologically more realistic and capable of handling complicated data, even spatiotemporal data. To exhibit its power and potential, we present the flexible transmitter network (FTNet), which is built on the most common fully connected feedforward architecture taking the FT model as the basic building block. FTNet allows gradient calculation and can be implemented by an improved backpropagation algorithm in the complex-valued domain. Experiments on a broad range of tasks show that FTNet has power and potential in processing spatiotemporal data. This study provides an alternative basic building block in neural networks and exhibits the feasibility of developing artificial neural networks with neuronal plasticity.

Clinicopathological significance of primitive phenotypes in early gastric cancer with differentiated histology.

BACKGROUND: Certain gastric cancers exhibit some primitive phenotypes, which may indicate a high malignancy. In histologically differentiated early gastric cancer (EGC), the presence and the clinicopathological significance of the primitive phenotype remain unclear. METHODS: Using immunohistochemical staining we detected the expression of three primitive phenotypic markers SALL4, Glypican-3(GPC3), and AFP in whole tissue sections of differentiated EGC (gastrectomy specimens, n = 302). For those cases with primitive phenotypes, we analyzed their clinicopathological features and evaluated whether the criteria for endoscopic resection were met. RESULTS: We found that 9.3% (28/302) of all differentiated EGC cases have primitive phenotypes, and most of these cases (25/28) exhibit a histomorphology similar to conventional differentiated EGC. Patients with primitive phenotypes had a deeper invasion, a higher rate of ulcer and lymphatic invasion than cases without primitive phenotype. Moreover, patients with primitive phenotypes displayed a significantly higher frequency of LNM than those without (57.1% vs 8.8%, P < 0.001). Multivariate analysis revealed that presence of primitive phenotypes was an independent risk factor for LNM (P = 0.001, HR 6.977, 95% CI: 2.199-22.138). Interestingly, we found 2 cases with primitive phenotypes developed LNM, and they both met the expanded indications of endoscopic resection for differentiated EGC. CONCLUSIONS: A small number of differentiated EGC have primitive phenotypes, which were closely related to LNM and were an independent risk factor for LNM. Given its highly aggressive behavior, differentiated EGC with primitive phenotypes should be evaluated with stricter criteria before endoscopic resection, or considered to give an additional surgical operation after endoscopic resection.

Persistence with medical treatment for Wilson disease in China based on a single center's survey research.

BACKGROUND: Wilson's disease (WD) is one of the few hereditary diseases that can be successfully treated with medicines. We conduct this survey research to assess treatment persistence among patients with WD and try to identify what factors affect the treatment persistence. METHODS: We employed WeChat which is the most popular social software in China to carry out this anonymous questionnaire research. The questionnaire included medication adherence scale. We also collected available medical records related to demographic and clinical characteristics. All the patients were divided into group of persistence with drug treatment (PDT) and nonpersistence with drug treatment (n-PDT). RESULTS: We collected 242 qualified questionnaires. Only 66.5% of patients were PDT during the mean 12.6 years of follow-up. In PDT group, better outcomes were observed: improvement (78.3%) and no change (16.1%) versus those in n-PDT (55.6%; and 28.4%, respectively). In PDT group, only nine patients deteriorated (6.8%) in comparison with 13 patients in n-PDT (16.0%). The adverse events (AEs) in PDT group were significantly less than those in n-PDT group. There were no significant differences in clinical type, gender, age, education level, and family knowledge about WD between the two groups. There were significant differences in AEs and family position toward treatment. CONCLUSION: Medication Adherence of Chinese WD patients was low. One third of the patients (33.5%) were unable to PDT, and it had an important negative effect on clinical outcome. AEs and family support had an important impact on treatment persistence.

Towards Safe Weakly Supervised Learning.

In this paper, we study weakly supervised learning where a large amount of data supervision is not accessible. This includes i) incomplete supervision, where only a small subset of labels is given, such as semi-supervised learning and domain adaptation; ii) inexact supervision, where only coarse-grained labels are given, such as multi-instance learning and iii) inaccurate supervision, where the given labels are not always ground-truth, such as label noise learning. Unlike supervised learning which typically achieves performance improvement with more labeled examples, weakly supervised learning may sometimes even degenerate performance with more weakly supervised data. Such deficiency seriously hinders the deployment of weakly supervised learning to real tasks. It is thus highly desired to study safe weakly supervised learning, which never seriously hurts performance. To this end, we present a generic ensemble learning scheme to derive a safe prediction by integrating multiple weakly supervised learners. We optimize the worst-case performance gain and lead to a maximin optimization. This brings multiple advantages to safe weakly supervised learning. First, for many commonly used convex loss functions in classification and regression, it is guaranteed to derive a safe prediction under a mild condition. Second, prior knowledge related to the weight of the base weakly supervised learners can be flexibly embedded. Third, it can be globally and efficiently addressed by simple convex quadratic or linear program. Finally, it is in an intuitive geometric interpretation with the least square loss. Extensive experiments on various weakly supervised learning tasks, including semi-supervised learning, domain adaptation, multi-instance learning and label noise learning demonstrate our effectiveness.

Heterogeneous Few-Shot Model Rectification With Semantic Mapping.

There still involve lots of challenges when applying machine learning algorithms in unknown environments, especially those with limited training data. To handle the data insufficiency and make a further step towards robust learning, we adopt the learnware notion Z.-H. Zhou, "Learnware: On the future of machine learning," Front. Comput. Sci., vol. 10, no. 4 pp. 589-590, 2016 which equips a model with an essential reusable property-the model learned in a related task could be easily adapted to the current data-scarce environment without data sharing. To this end, we propose the REctiFy via heterOgeneous pRedictor Mapping (ReForm) framework enabling the current model to take advantage of a related model from two kinds of heterogeneous environment, i.e., either with different sets of features or labels. By Encoding Meta InformaTion (Emit) of features and labels as the model specification, we utilize an optimal transported semantic mapping to characterize and bridge the environment changes. After fine-tuning over a few labeled examples through a biased regularization objective, the transformed heterogeneous model adapts to the current task efficiently. We apply ReForm over both synthetic and real-world tasks such as few-shot image classification with either learned or pre-defined specifications. Experimental results validate the effectiveness and practical utility of the proposed ReForm framework.

SHARPIN stabilizes ß-catenin through a linear ubiquitination-independent manner to support gastric tumorigenesis.

BACKGROUND: Aberrant activation of Wnt/ß-catenin signaling by dysregulated post-translational protein modifications, especially ubiquitination is causally linked to cancer development and progression. Although Lys48-linked ubiquitination is known to regulate Wnt/ß-catenin signaling, it remains largely obscure how other types of ubiquitination, such as linear ubiquitination governs its signaling activity. METHODS: The expression and regulatory mechanism of linear ubiquitin chain assembly complex (LUBAC) on Wnt/ß-catenin signaling was examined by immunoprecipitation, western blot and immunohistochemical staining. The ubiquitination status of ß-catenin was detected by ubiquitination assay. The impacts of SHARPIN, a core component of LUBAC on malignant behaviors of gastric cancer cells were determined by various functional assays in vitro and in vivo. RESULTS: Unlike a canonical role in promoting linear ubiquitination, SHARPIN specifically interacts with ß-catenin to maintain its protein stability. Mechanistically, SHARPIN competes with the E3 ubiquitin ligase ß-Trcp1 for ß-catenin binding, thereby decreasing ß-catenin ubiquitination levels to abolish its proteasomal degradation. Importantly, SHARPIN is required for invasiveness and malignant growth of gastric cancer cells in vitro and in vivo, a function that is largely dependent on its binding partner ß-catenin. In line with these findings, elevated expression of SHARPIN in gastric cancer tissues is associated with disease malignancy and correlates with ß-catenin expression levels. CONCLUSIONS: Our findings reveal a novel molecular link connecting linear ubiquitination machinery and Wnt/ß-catenin signaling via SHARPIN-mediated stabilization of ß-catenin. Targeting the linear ubiquitination-independent function of SHARPIN could be exploited to inhibit the hyperactive ß-catenin signaling in a subset of human gastric cancers.