arXiv subjects

Ziwei Yang

Publications and source records attributed to Ziwei Yang.

27 records · Page 2Linked to original sources

arXiv2022

Automated Cancer Subtyping via Vector Quantization Mutual Information Maximization

Cancer subtyping is crucial for understanding the nature of tumors and providing suitable therapy. However, existing labelling methods are medically controversial, and have driven the process of subtyping away from teaching signals. Moreover, cancer genetic expression profiles are high-dimensional, scarce, and have complicated dependence, thereby posing a serious challenge to existing subtyping models for outputting sensible clustering. In this study, we propose a novel clustering method for exploiting genetic expression profiles and distinguishing subtypes in an unsupervised manner. The proposed method adaptively learns categorical correspondence from latent representations of expression profiles to the subtypes output by the model. By maximizing the problem -- agnostic mutual information between input expression profiles and output subtypes, our method can automatically decide a suitable number of subtypes. Through experiments, we demonstrate that our proposed method can refine existing controversial labels, and, by further medical analysis, this refinement is proven to have a high correlation with cancer survival rates.

cs.LG↗

arXiv2022

Cancer Subtyping by Improved Transcriptomic Features Using Vector Quantized Variational Autoencoder

Defining and separating cancer subtypes is essential for facilitating personalized therapy modality and prognosis of patients. The definition of subtypes has been constantly recalibrated as a result of our deepened understanding. During this recalibration, researchers often rely on clustering of cancer data to provide an intuitive visual reference that could reveal the intrinsic characteristics of subtypes. The data being clustered are often omics data such as transcriptomics that have strong correlations to the underlying biological mechanism. However, while existing studies have shown promising results, they suffer from issues associated with omics data: sample scarcity and high dimensionality. As such, existing methods often impose unrealistic assumptions to extract useful features from the data while avoiding overfitting to spurious correlations. In this paper, we propose to leverage a recent strong generative model, Vector Quantized Variational AutoEncoder (VQ-VAE), to tackle the data issues and extract informative latent features that are crucial to the quality of subsequent clustering by retaining only information relevant to reconstructing the input. VQ-VAE does not impose strict assumptions and hence its latent features are better representations of the input, capable of yielding superior clustering performance with any mainstream clustering method. Extensive experiments and medical analysis on multiple datasets comprising 10 distinct cancers demonstrate the VQ-VAE clustering results can significantly and robustly improve prognosis over prevalent subtyping systems.

cs.LG↗

arXiv2022

Multi-Tier Platform for Cognizing Massive Electroencephalogram

An end-to-end platform assembling multiple tiers is built for precisely cognizing brain activities. Being fed massive electroencephalogram (EEG) data, the time-frequency spectrograms are conventionally projected into the episode-wise feature matrices (seen as tier-1). A spiking neural network (SNN) based tier is designed to distill the principle information in terms of spike-streams from the rare features, which maintains the temporal implication in the nature of EEGs. The proposed tier-3 transposes time- and space-domain of spike patterns from the SNN; and feeds the transposed pattern-matrices into an artificial neural network (ANN, Transformer specifically) known as tier-4, where a special spanning topology is proposed to match the two-dimensional input form. In this manner, cognition such as classification is conducted with high accuracy. For proof-of-concept, the sleep stage scoring problem is demonstrated by introducing multiple EEG datasets with the largest comprising 42,560 hours recorded from 5,793 subjects. From experiment results, our platform achieves the general cognition overall accuracy of 87% by leveraging sole EEG, which is 2% superior to the state-of-the-art. Moreover, our developed multi-tier methodology offers visible and graphical interpretations of the temporal characteristics of EEG by identifying the critical episodes, which is demanded in neurodynamics but hardly appears in conventional cognition scenarios.

eess.SP↗

arXiv2022

Cancer Subtyping via Embedded Unsupervised Learning on Transcriptomics Data

Cancer is one of the deadliest diseases worldwide. Accurate diagnosis and classification of cancer subtypes are indispensable for effective clinical treatment. Promising results on automatic cancer subtyping systems have been published recently with the emergence of various deep learning methods. However, such automatic systems often overfit the data due to the high dimensionality and scarcity. In this paper, we propose to investigate automatic subtyping from an unsupervised learning perspective by directly constructing the underlying data distribution itself, hence sufficient data can be generated to alleviate the issue of overfitting. Specifically, we bypass the strong Gaussianity assumption that typically exists but fails in the unsupervised learning subtyping literature due to small-sized samples by vector quantization. Our proposed method better captures the latent space features and models the cancer subtype manifestation on a molecular basis, as demonstrated by the extensive experimental results.

cs.LG↗

arXiv2022

Adaptive Spike-Like Representation of EEG Signals for Sleep Stages Scoring

Recently there has seen promising results on automatic stage scoring by extracting spatio-temporal features from electroencephalogram (EEG). Such methods entail laborious manual feature engineering and domain knowledge. In this study, we propose an adaptive scheme to probabilistically encode, filter and accumulate the input signals and weight the resultant features by the half-Gaussian probabilities of signal intensities. The adaptive representations are subsequently fed into a transformer model to automatically mine the relevance between features and corresponding stages. Extensive experiments on the largest public dataset against state-of-the-art methods validate the effectiveness of our proposed method and reveal promising future directions.

eess.SP↗

arXiv2022

Monolithic Silicon Carbide Metalenses

Silicon carbide has emerged as a promising material platform for quantum photonics and nonlinear optics. These properties make the development of integrated photonic components in high-quality silicon carbide a critical aspect for the advancement of scalable on-chip networks. In this work, we numerically design, fabricate and demonstrate the performance of monolithic metalenses from SiC suitable for on-chip optical operations. We engineer two distinct lenses, with parabolic and cubic phase profiles, operating in the near infrared spectral range, which is of interest for quantum and photonic applications. We support the lens fabrication by optical transmission measurement and characterize the focal points of the lenses. Our results will accelerate the development of SiC nanophotonic devices and aid in an on chip integration of quantum emitters with meta-optical components.

physics.optics↗

arXiv2021

Cascade Bagging for Accuracy Prediction with Few Training Samples

Accuracy predictor is trained to predict the validation accuracy of an network from its architecture encoding. It can effectively assist in designing networks and improving Neural Architecture Search(NAS) efficiency. However, a high-performance predictor depends on adequate trainning samples, which requires unaffordable computation overhead. To alleviate this problem, we propose a novel framework to train an accuracy predictor under few training samples. The framework consists ofdata augmentation methods and an ensemble learning algorithm. The data augmentation methods calibrate weak labels and inject noise to feature space. The ensemble learning algorithm, termed cascade bagging, trains two-level models by sampling data and features. In the end, the advantages of above methods are proved in the Performance Prediciton Track of CVPR2021 1st Lightweight NAS Challenge. Our code is made public at: https://github.com/dlongry/Solutionto-CVPR2021-NAS-Track2.

cs.CV↗

arXiv2021

Improving Ranking Correlation of Supernet with Candidates Enhancement and Progressive Training

One-shot neural architecture search (NAS) applies weight-sharing supernet to reduce the unaffordable computation overhead of automated architecture designing. However, the weight-sharing technique worsens the ranking consistency of performance due to the interferences between different candidate networks. To address this issue, we propose a candidates enhancement method and progressive training pipeline to improve the ranking correlation of supernet. Specifically, we carefully redesign the sub-networks in the supernet and map the original supernet to a new one of high capacity. In addition, we gradually add narrow branches of supernet to reduce the degree of weight sharing which effectively alleviates the mutual interference between sub-networks. Finally, our method ranks the 1st place in the Supernet Track of CVPR2021 1st Lightweight NAS Challenge.

cs.CV↗

arXiv2020

Modulating nano-inhomogeneity at electrode-solid electrolyte interfaces for dendrite-proof solid-state batteries and long-life memristors

The penetration of dendrites in ceramic lithium conductors severely constrains the development of solid-state batteries (SSBs) while its nanoscopic origin remain unelucidated. We develop an in-situ nanoscale electrochemical characterization technique to reveal the nanoscopic lithium dendrite growth kinetics and use it as a guiding tool to unlock the design of interfaces for dendrite-proof SSBs. Using Li7La3Zr2O12 (LLZO) as a model system, in-situ nanoscopic dendrite triggering measurements, ex-situ electro-mechanical characterizations, and finite element simulations are carried out which reveal the dominating role of Li+ flux detouring and nano-mechanical inhomogeneity on dendrite penetration. To mitigate such nano-inhomogeneity, an ionic-conductive homogenizing layer based on poly(propylene carbonate) is designed which in-situ reacts with lithium to form a highly conformal interphase at mild conditions. A high critical current density of 1.8mA cm-2 and a low interfacial resistance of 14Ω cm2 is achieved. Practical SSBs based on LiFePO4 cathodes show great cyclic stability without capacity decay over 300 cycles. Beyond this, highly reversible electrochemical dendrite healing behavior in LLZO is discovered using the nano-electrode, based on which a model memristor with a high on/off ratio of ~10^5 is demonstrated for >200 cycles. This work not only provides a novel tool to investigate and design interfaces in SSBs but offers also new opportunities for solid electrolytes beyond energy applications.

physics.app-ph↗